Animal geography : the faunas of the natural regions of the globe

Gurvell University Library
Ithaca, New York

BOUGHT WITH THE INCOME OF THE

SAGE ENDOWMENT FUND

THE GIFT OF

HENRY W. SAGE

1891

aaah seqet University Library

win

24

Cornell University

Library

The original of this book is in
the Cornell University Library.

There are no known copyright restrictions in
the United States on the use of the text.

http://Awww.archive.org/details/cu31924024771515

THE OXFORD GEOGRAPHIES
Epirep sy A. J. HERBERTSON

ANIMAL GEOGRAPHY

THE FAUNAS OF THE NATURAL
REGIONS OF THE GLOBE

BY

MARION I. NEWBIGIN
D.Sc. (Lonp.)

LECTURER IN ZOOLOGY AND BIOLOGY IN THE EDINBURGH SCHOOL OF
MEDICINE FOR WOMEN
EDITOR OF THE ‘ SCOTTISH GEOGRAPHICAL MAGAZINE’

OXFORD
AT THE CLARENDON PRESS

1913
$

j
\.37095 + |
OXFORD UNIVERSITY PRESS
LONDON EDINBURGH GLASGOW NEW YORK
TORONTO MELBOURNE BOMBAY
HUMPHREY MILFORD M.A.

PUBLISHER TO THE UNIVERSITY

PREFACE

WITHIN the last few years, as is well known, botanical
geography has made great strides. The intimate rela-
tion between the vegetative covering and the physical
conditions has been in many cases thoroughly worked
out, and all the more modern geographical text-books
give at least some indication of the results obtained.
On the other hand, the more difficult subject of the
relation between the animals of the various natural
regions and their surroundings has been much less
fully treated by geographers. The facts have still
mostly to be sought in zoological text-books, where
they are treated from another point of view, and many
even of the most recent geographical text-books show
uncertainty of treatment when dealing with the
animals of tropical forest and arctic tundra. The
present book is an attempt to put the main facts
connected with the distribution of animals in a form
acceptable to the geographical student, and has been
written by one whose approach to geography was first
made from the biological side.

The sources of the illustrations are indicated beneath
each, and I am greatly indebted to the various
gentlemen named for permission to reproduce their
photographs. Special acknowledgement should be
made to Sir Thomas Carlaw Martin, Director of

4 PREFACE

the Royal Scottish Museum, and Mr. Eagle Clarke,
Keeper of the Natural History Collection there, for
permission to photograph a number of specimens in
the collection.

For the index I am indebted to my sister, Miss

Florence Newbigin.
MARION I. NEWBIGIN.

Epinsuresu, 1913.

CONTENTS

PAGE
Intropuction: THe NaturaL REGIONS OF THE GLOBE . 9
CHAPTER

I. Tar Tunpra anp its Fauna P . 15

II. Tae Tatca, oR ConrFEROUS FoREST, AND ITS Fauna. 36

III. Steprz Faunas AND THE TEMPERATE STEPPES OF

Asta AND NortH AMERICA . 5 . 62

IV. Movntarn Faunas : - , é - 72
V. Tue Fauna or THe TROPICAL ForEST . : . 92
VI. Tropica Savanas AND DESERTS “ ‘ . 121
VII. Spsctat Features or Istanp Faunas: . ; . 144
VIII. THe Distrrution of Anmm4L LirE IN THE SEA . 156

IX. THE Anrats oF Lakes AnD Rivers. CAvE Faunas 182
X. ZoOOGEOGRAPHICAL REGIONS . ‘ - : . 208

APPENDIX. OUTLINE CLASSIFICATION OF ANIMALS. . 226

INDEX . : 2 r ‘ 3 ‘ F F . 231

FIG.
. Tundra in Northern Russia

bo TO NO oe ll el el sel
ee NN ON Se SCOOP AME SHBNBDNAAR WHY

ie)
rq

LIST OF ILLUSTRATIONS

Ptarmigan .

. Wapiti

Elk

. Virginian Deer

. Three-toed Woodpecker
. Camels :

. American Bison .

. Prairie Marmot .

. Sage-cock .

. Swiss Alp .

. Snow Leopard

. Yak . ‘i - F

. Alaskan Wild Sheep .

. Takin <

. Kiang

. Bharal

. Serow

. Musk-deer .

. Hyrax é . : .
. Rain-forest in North Borneo
. Congo Tropical Forest

. Red Howling Monkey

. American Monkey

. Black Lemur

. Puma and Cubs

. African Elephant

. Indian Elephant

. African Flying Squirrel

. Brazilian Tree Porcupine
. Quica Opossum .

. Sloth

. Phalanger .

. Savana

100
101
102
108
109
109
110
lll
ill
112.
113
122

8 LIST OF ILLUSTRATIONS

via. PAGE
35. Gnu . : F F : ; facing 130
36. Giraffe ¥ 3 : 3 99 131
37. Jerboa (Central Asia) . : 7 . 56 134
38. Jerboa (Egypt) . , é ‘ ; ‘ * 134
39. Cactus Wren ; ; ; : : : . 141
40. Kiwi . . ; ; . Jacing 150
41. Seal Rookery in ‘Copper Island : ; : 3 162
42. Seals on shore of Copper Island . j é be 163
43. Kittiwake Gulls . ; : ; : , , . 164
44. Penguins . i F F ‘i F facing 164
45. Clouded Leopard : ‘ i : ‘ ah 218
46. Okapi 2 , ‘ é r 3 7 35 219
47. Bower Bird i ; : 2 ‘ ss 224

225

48. Lyre Bird . 4 ; é ‘ : P 55

INTRODUCTION

THE NATURAL REGIONS OF THE GLOBE

THE modern interest in the facts of the distribution
of animals, as indeed the modern interest in so many
of the problems connected with biology, dates from
the publication of the Origin of Species. Darwin in his
famous voyage on the Beagle was much struck by the
curious phenomena of distribution which that voyage
brought before his eyes, and it was these phenomena
which helped to direct his attention to that line of
thought which resulted ultimately in the publication
of his theory of Natural Selection. The two facts which
especially struck him were the two set forth in the
chapters on distribution in the Origin of Species. These
are, first, that similarity of physical conditions does not
necessarily result in a similar fauna, and conversely.
For example, there is considerable uniformity in the
fauna of South America throughout its length and
breadth, even though we find there every variety of
physical conditions, from the cold deserts of the high
Andes to the luxuriance of the Brazilian forest, or from
the burning deserts of Chile to the grassy plains of the
Argentine. On the other hand, there is very little
resemblance between the animals of the grassy plains
of South America and those of Africa, despite the
similarity in physical conditions. The second point
which struck him was that the resemblances or differ-
ences between the faunas of two areas are directly

10 INTRODUCTION

proportioned to the nature of the barrier which sepa-
rates the two regions. Transverse mountain chains,
such as the Himalayas or Atlas, separate two different
faunas. Similarly, lands separated by a broad and
deep belt of ocean may have quite different faunas ;
those separated by a shallow and narrow strait may
have almost identical animals. Such facts led him to
believe that each species originated in a particular
region, and spread out from that region till it was
stopped by some barrier to further distribution. The
remarkable features of the animals of isolated regions
can be explained on this hypothesis as the result of
descent with modification from the original forms.

One result of the direction thus given to the study of
distribution has been to make the division of the globe
into zoogeographical regions, and the study of the faunas
of the different regions, of supreme importance, and it
is this aspect which is studied in most books on the
subject. But, obviously, another and more truly
geographical point of view is possible. We may study
the animals of the natural regions of the globe as form-
ing a part of the features of those regions, and in this
case attention is directed rather to the adaptations
displayed by the animals, than to their zoological
relationships.

In deliminating the natural regions of the globe the
botanists have played a large part, for they have shown
that the existence of the great plant associations
depends upon the climate, the relief, and so forth of
the region. To the practised eye a collection of plants
from e.g. the Mediterranean region gives a great deal
of information in regard to the region. For example,
the small silvery leaves of such plants, and the other
devices for husbanding water, together with the occur-

NATURAL REGIONS OF THE GLOBE 11

rence of many bulbous plants, speaks to a climate where
dry and wet seasons alternate, and so on. The depen-
dence of the more active animal upon its surroundings is
less intimate than that of the passive plant, but even
here the characters of the animals of a region tell us
something of the physical features of that region. That
we find in Brazil, for example, animals so thoroughly
adapted to the arboreal life as sloths, marmosets,
monkeys, and so forth, suggests at once that dense
forest occurs in the region. That Africa has an
enormous number of antelopes—enormous both as
regards species and individuals—suggests the existence
in that continent of open plains, and so with other
regions.

The botanists recognize three great plant formations,
which roughly divide the surface of the land among
them. These are woodland, grassland, and desert.
Such a simple division would help us little in the case
of animals, but as all animals depend ultimately for
their food upon plants, a modification of this classi-
fication may serve as a framework in the study of
animal distribution. There is, for example, a consider-
able difference in the life conditions of animals which
live in regions where the winter cold periodically checks
plant growth, as compared with those where no such
temperature check occurs. This gives us a first dis-
tinction between the animals of cold or temperate
regions and those of tropical regions. In both of these
forest, grassland, and desert occur, but it is possible to
pick out certain well-defined types in order to study
their faunas separately. Thus, among the various forms
of temperate forest, the Taiga or coniferous forest of
Asia and North America has a well-defined assemblage
of animals showing some interesting adaptations.

12 INTRODUCTION

Again, both in Asia and in North America we have
vast areas of natural grassland forming well-defined
Steppe areas, which possess a characteristic fauna. The
great mountain chains of the earth’s surface are peopled
by faunas having certain special features, and the vast
area of coast-line which fringes the Arctic sea forms
a cold desert or tundra, whose fauna has many pecu-
liar features. As natural regions with a cold winter
climate then, we have to consider (1) the Tundra,
(2) the Taiga or coniferous forest, (3) the Steppe, best
developed in the northern hemisphere, and (4) Mountain
regions.

Where the temperature is permanently high, and
the precipitation great, we have (5) luxuriant Tropical
Forest, with its characteristic fauna, differing according
as the forest occurs in Africa, Asia, or South America,
but showing in all cases certain special adaptations.
Again, in warm or tropical regions where the tempera-
ture is still high, but the rainfall small and unequally
distributed throughout the year, we have a thinning of
the forest into that type of grassland, or grassland
mingled with wood, which the botanists call Savana.
As the rainfall diminishes the savana thins out into
desert, so that we may conveniently consider (6)
Savanas and Warm Deserts together.

This still leaves to be considered the fauna of (7)
Islands, which show some special features, and also the
animals of aquatic habitat, which we must consider
separately, according as they live (8) in the sea, or
(9) in fresh water.

Such a scheme is obviously incomplete, in that it
does not cover the whole surface of the globe. It
omits, for example, any consideration of the fauna of
the deciduous forest which covers so much ground in

NATURAL REGIONS OF THE GLOBE 13

Europe and North America, and of the animals of the
Mediterranean or scrub forest, found round the Mediter-
ranean Sea, in California, and elsewhere. We shall find,
however, that such regions have for the most part
a transitional fauna, made up of units from the sur-
rounding regions, and that further they have been
much altered by man’s interference. It is true that the
effects of this interference are often interesting, for
while some animals die out as civilization spreads,
others, like the rats, the common sparrow, the cock-
roaches, some parasites, and so on, adjust themselves
to the altered conditions, and prosper under civiliza-
tion as they never did in former days. But a considera-
tion of these points would take us too far from our
immediate subject. We shall therefore confine our-
selves to a consideration of the nine natural regions
already given, and fill up whatever gaps this method
may leave by a final consideration of the zoogeogra-
phical regions into which zoologists have divided the
globe. But as the greater part of this book is thus
devoted to the geographical aspect of animal distribu-
tion, a word or two may be added in further explanation
of the difference between the two points of view, the
zoological and the geographical.

To the zoologist it is a fact of great interest
that the sloths, ant-eaters, and armadillos should
occur in South America, and nowhere else. But the
geographer is more interested in the fact that the sloth
is a purely arboreal animal, fitted only for life in the
dense forest, while the armadillo, for example, shows
adaptations to quite other habitats. The difference
between the sloth of the Amazon forest and the squirrel
of the Siberian taiga, again, means something very
different to zoologist and geographer, for the latter

14 INTRODUCTION

sees in it a reflection of the difference between the
two forests, which interests the other less, and only
incidentally.

Rererences. A realization of the connexion between the study of
the facts of animal distribution and the general theory of evolution may
best be obtained from Darwin’s Origin of Species and Wallace’s Dar-
winism. Wallace’s classical work on Geographical Distribution (two vols.,
London, 1876) should also be consulted. An interesting discussion of
the purely geographical standpoint will be found in the chapters on
animal geography in de Martonne’s T'raité de Géographie Physique (Paris,
1909), which gives copious references.

CHAPTER I
THE TUNDRA AND ITS FAUNA

One of the well-marked natural regions of the world
is the Tundra, that treeless area which fringes the
margin of the far northern seas, and is characterized
by its climate, its plants and animals, and to some
extent also by its topography. Originally applied only
to the treeless lands of Northern Asia, lying to the
north of the forest or taiga, the term is now by exten-
sion given to all areas of similar character, occurring
alike in the Old and New Worlds. In the southern
hemisphere the tapering of the land-masses, and their
cessation in relatively low latitudes, make the develop-
ment of true tundra impossible. It is true that con-
siderable land-masses occur within the Antarctic area, in
latitudes lower than those in which the northern tundra
finds its limit, but the climate and the absence of a direct
connexion between these areas and the continents have
prevented the development of the characteristic tundra
plants and animals. In the Antarctic land-mass flower-
ing plants are, with two exceptions, absent, and mosses
and lichens are only slightly developed. Associated
with this we have a complete absence of land mammals
—a very striking difference from the northern tundra.

The boundaries of the tundra are well marked. As
already indicated, on the continents the southern
boundary is the beginning of the forest zone. This
limit is, roughly speaking, marked out by the July
isotherm of 50°. Wherever the mean temperature of
the hottest month exceeds 50° F., there the tundra

v

16 THE TUNDRA AND ITS FAUNA

dies away into forest. The result is that the continental
tundra forms a band of very varying width. The mild-
ness of the Western European climate pushes the tree
limit up within the Arctic Circle, and reduces the tundra
zone to a narrow band to the north of Scandinavia.
Thence it extends through European Russia into
Siberia, and reappears again at the other side of the
Bering Strait to extend across Arctic America. Here
to the east it revenges itself for its conquest by the
forest in Western Europe by pushing the forest zone
far to the south, so that a band of tundra extends
down the coast-line of Labrador, and even into New-
foundland.

The northern limit of the continental tundra is the
polar sea, but the tundra type reappears on the margin
of the land-masses lying within that sea. The islands
of the American archipelago, Baffin Land, Greenland,
the northern shore of Iceland, Spitsbergen, Franz
Josef Land, Nova Zembla, the New Siberian Islands,
&c., are all fringed by a band of tundra. Its northern
extension may indeed be gathered from the fact that
the reindeer, one of the most characteristic tundra
animals, was hunted by the Sverdrup expedition in
King Oscar Land, which extends to about 80° 30’
N. lat. In these regions the sea again forms one of the
boundaries of the tundra zone. The other boundary is
formed by the presence of eternal ice and snow. Gener-
ally it may be said that tundra occurs wherever the
melting of the snow is sufficient to expose a surface
upon which land plants may grow.

The climate throughout this great area naturally
varies, but the special features are the following.
Nowhere does the mean temperature of the warmest
month exceed 50° F., and it is often much below this.

THE TUNDRA AND ITS FAUNA 17

The precipitation is everywhere small, except in the
south-western extremity of Greenland. Elsewhere the
mean never exceeds 10’’ per annum, and may be much
less. The winters especially are remarkable for the
clearness and dryness of the atmosphere and the
virtual absence of precipitation. The result is that,
contrary to what might be expected, the snow does not
form everywhere a thick mantle, and the frequency of
strong winds helps to diminish its thickness in places,
a fact which greatly aids the obtaining of food by
such herbivorous mammals as the reindeer and musk-
ox. During the short summer precipitation is frequent,
though slight, and fog and mist are common. The sub-
soil is permanently frozen, and there is thus no ground
water. This renders all erosion mechanical, and as
springs are absent the rivers are fed only by the melt-
ing of ice and snow, and are temporary, being frost-
bound in winter.

There are only two seasons—winter and summer—
and the winter extends far into the months which are
spring elsewhere, so that March or even April may be
the coldest month. It is apparently these months which
take the largest toll of animal life, for Nordenskiéld
notes that the reindeer are in better condition in the
depth of winter than in the spring months. Summer
comes with extraordinary rapidity, and with its coming
life suddenly bursts into activity. Thus all the flowers
bloom at once, giving rise to the local luxuriance
noticed by many explorers. The rapidity of develop-
ment is of course aided by the virtually continuous
daylight. As the rays of sunlight have a very oblique
direction, local conditions count for much in promoting
the growth of plants, and therefore the occurrence of
land animals. Thus a slope where the ice-cold water

1404 B

18 THE TUNDRA AND ITS FAUNA

drains away as the snow melts is favourable to plant
growth, while level ground is very unfavourable. A
sunny exposure also exerts much influence. The Sver-
drup expedition noted that the Arctic hare is prac-
tically limited to the regions where large valleys open
out into fiords, for only here is vegetation abundant
enough to supply its needs.

As regards topography one must note that the tundra,
like the semi-arid lands of other latitudes, is charac-
terized by a very youthful condition of drainage. As
already stated, the rivers are temporary, chemical
erosion does not occur, and therefore the process of
carving up the land into hill and valley goes on very
slowly, and the land is imperfectly drained. ‘The
imperfect drainage is aided by the fact that the ex-
treme coldness of the soil is unfavourable to bacterial
life, and thus humus accumulates and favours the
growth of bog mosses as contrasted with higher plants.
These hold up the water like a sponge, and so hinder
drainage. When Nansen and his party left the inland
ice of Greenland they descended to a little tarn, whose
water drained by a stream to the distant fiord. In
their descent, as the leader notices, they found that the
main streams had often transverse tributaries, which
greatly hindered progress. The picture may be taken
as characteristic of the tundra. Lakes fed by melting
ice, streams showing a very undeveloped drainage
system—these are characteristic features, which, aided
by the climate, account for the very local development
of a covering of vegetation, and therefore for the
curiously local appearance of animals.

Without discussing the plants of the tundra in detail
we may note those points which are of special impor-
tance for animal life. The first of these is the great

THE TUNDRA AND ITS FAUNA 19

abundance of mosses and lichens, which do not die
down in winter, and are therefore permanently avail-
able as pasture for the herbivorous mammals. It is
well known that the lichen popularly called reindeer
moss (Cladonia rangiferina), which occurs over wide
areas, forms a very important part of the winter food

Fic. 1. The Tundra in Northern Russia, summer conditions. The
illustration shows the undifferentiated nature of the drainage, and
shows also that at this season pasturage is freely available for
herbivorous mammals.

of the reindeer. Other areas are clad in hair moss
(Polytrichum), or bog moss (Sphagnum). Among the
higher plants we may note the presence of dwarf
willows, birches, alders, &c., whose branches stick up
through the snow, and whose buds and seeds are of
great importance as part of the food of the ptarmigan
and the snow-bunting. In the more favourably situated
regions berry-bearing shrubs occur, such as crowberries,
B2

20 THE TUNDRA AND ITS FAUNA

whortleberries, &c., again forming an important part of
the available food supply. Again, as in the case of the
Swiss alps (p. 74), the vegetative growth of the plants
is checked by the natural conditions, and thus, though
stunted in appearance, they are relatively rich in
proteids and fats, and relatively poor in cellulose, which
makes their nutritive value high. With all that can
be said, however, the land offers but little food, and.
the vast majority of the tundra animals depend partly
or wholly upon the sea, and are therefore necessarily
confined in their distribution to the margin of the sea.
The German expedition to Greenland made large
collections of both birds and mammals, and observed
the habits of all as closely as possible, and the conclu-
sion arrived at is that only four of the mammals can
be said to be permanently dependent upon the land
for food ; these four being the reindeer, the musk-ox,
the Arctic hare, and the lemming. Of the many birds
known to visit Greenland, only two can be said to be
at once permanent residents and wholly dependent
upon the land for food. These are the redpoll (Linota
hornemanni) and the ptarmigan (Lagopus hyperboreus).
All other birds and mammals are either partially
migratory or depend to some extent at least upon the
harvest of the sea. Even of the land plants the last
statement is partly true, for it is said that the phanero-
gams are especially abundant about the breeding-places
of the sea birds, on ground enriched by their droppings,
and such birds feed in the sea. Again, even of the four
mammals named, the reindeer, according to Norden-
skidld, does not disdain seaweed in times of scarcity.
Everywhere over the globe animals depend ultimately
upon plants, which alone can fix inorganic matter
in organic form, but it may be’ said to be the special

THE TUNDRA AND ITS FAUNA 21

character of the tundra animals, like the fewer animals
of the Antarctic area, that it is marine plants, rather
than terrestrial ones, which form the basis of their
food supply.

Having thus summarized the conditions of life in the
tundra we may give some account of the fauna, begin-
ning with the ungulates among mammals. The rein-
deer (Rangifer tarandus) is widely spread throughout.
the region, though nowhere very abundant as indi-
viduals. On the continent of America there is a marked
distinction between those forms which haunt the forests,
the so-called woodland caribou, and the smaller forms,
with larger antlers, which occur in the tundra. The
latter, the barren-ground forms, seem to migrate south-
wards towards the forest in winter, but this is impossible
with the herds which occur in the Arctic islands. In
King Oscar Land and the adjacent regions the Sverdrup
expedition found the animal widely distributed but not
abundant, apparently on account of persecution by the
Arctic wolf. The absence of this animal in Spits-
bergen perhaps accounts for the greater abundance of
the reindeer there. It occurs also in the suitable parts
of Greenland, where the wolf is again absent. On the
continent of Asia the reindeer occurs in summer in
suitable localities in the tundra region, but here also it
seems migratory, and thus not wholly dependent upon
the tundra for food.

The musk-ox (Ovibos moschatus), the second large
herbivore of polar regions, has a much more limited
distribution. Found fossil in Europe and Asia, it is
now limited to the American side of Arctic regions.
On the continent of America its eastward distribution
is limited by the Mackenzie River, but it is apparently
abundant in the northern parts of Greenland, in

22 THE TUNDRA AND ITS FAUNA

Grinnell Land, and in the adjacent regions. It is
non-migratory, finding food throughout the year by
scraping away the thin covering of snow. A markedly
social animal, the herds seem capable of defending.
themselves against the Arctic wolf, which attacks
chiefly young stragglers.

To this short list we should perhaps add the fact that
the mammoth occurs in the tundra in the subfossil
condition in Asia and also in Alaska.

Rodents are represented in the tundra by the Arctic
hare (Lepus glacialis), which is very abundant in suit-
able localities. Even in winter it finds food enough in
the grasses which project through the snow. Its
greatest enemy is the wolf, which seems to feed largely
upon it, but in spite of the abundance of that animal
in King Oscar Land the Arctic hare is stated to be very
abundant there. Another important rodent of the
tundra is the lemming (Myodes torquatus), which is
sometimes extraordinarily abundant. Absent from
Spitsbergen, it occurs elsewhere on both sides of the
Arctic region. Perhaps because of the paucity of
vegetation in its natural habitat, this northern lemming
does not show the enormous fertility of the Norwegian
form, nor does it seem to migrate in the same fashion.
In some regions, as in Nova Zembla, it is, however,
remarkably abundant. While the other herbivorous
animals named find their food in winter either by
scraping away the snow, or by nibbling the protruding
shoots, the lemming, a much smaller animal—it is not
much bigger than a mouse—lives in winter beneath the
snow, making runs and burrows in the underlying
ground. When the snow melts in spring these runs
appear, ramifying over the surface in all directions.
a Norwegian lemming (IM. lemmus) is also circum-

THE TUNDRA AND ITS FAUNA 23

polar, but does not extend so far to the north as the
other species.

In the tundra, as elsewhere, the presence of large
herbivores naturally attracts their enemies the large
carnivores, necessarily fewer in number, and usually of
greater intelligence. As masterful and highly organized
animals the large carnivores do not as a rule show
a very close adaptation to one type of country only,
nor are they, as a rule, like such herbivorous animals
as the musk-ox, driven by stress of competition to the
unfavourable parts of the surface. The carnivores
which occur in the tundra, therefore, do not as a rule
form an integral part of its fauna; rather are they
intruders from other regions. Thus the wolf (Canis
lupus), absent in Greenland, in Spitsbergen, &c., occurs
in Arctic Asia and Arctic America, and extends also
into such regions as Grinnell Land and King Oscar
Land. Its near ally, the Eskimo dog, is a very impor-
tant animal in Greenland and elsewhere. Very in-
sufficiently fed by its master, the catholicity of its
appetite may be gathered from the fact that it will eat
fish, shell-fish, seaweed, and refuse.

Much more truly Arctic than the wolf is the Arctic
fox (Canis lagopus), widely distributed in the polar
regions of both the Old and New Worlds. Its food con-
sists of birds, especially the ptarmigan, but it is con-
strained to add to this refuse and even seaweed.
Lemmings are also an important part of its diet where
they occur, and by some it is believed to store the
bodies of these for the winter, when it is often hard-
pressed. Two other land carnivores occur in the tundra,
but both are immigrants from their natural home
further south. These are the ermine (Mustela erminea),
which follows the lemming to the coasts of the Arctic

24 THE TUNDRA AND ITS FAUNA

sea, and the glutton (Gulo luscus), a forest mammal,
which just reaches the tundra.

Among the mammals which find their food in the sea,
we must place the polar bear. As its name (Ursus mari-
timus) indicates, this animal haunts the margin of the
sea, where it is found among the drift and floe ice, on
the look out for its natural prey, seals and the walrus.
It is thus not in any true sense a tundra animal, al-.
though occurring round the margin of the tundra region
throughout its extension. Of the many kinds of seals.
also, we can only say here that though they breed on
the shore they are not in any sense tundra animals.

We may pass next to the birds of the tundra. In
regard to these it is noticeable that large numbers of
migrants occur, especially sea birds and those which.
haunt the vicinity of the shore or of fresh water.
These come north to breed in the relative security of
the tundra, and also probably because they find the
long northern daylight an advantage, for it lengthens
the period during which the search for food can be
carried on. As winter approaches, these birds travel
southwards, fleeing from the increasing cold and dark-
ness. It is unnecessary to give lists of these migratory
forms. All travellers to tundra regions, whether to the
mainland areas or to the lands of the polar sea, speak
of many kinds of gulls, of skuas, of kittiwakes, of
geese, of various forms of ducks, including the eider, of
auks and guillemots, of sandpipers, terns, petrels, and
so forth, none of which can be regarded as definitely
characteristic of the tundra, though their summer
abundance at certain special localities is one of the
features of the region. It is otherwise with the true
land birds, which, though relatively few in number,
help to give its characteristic facies to the region. First

THE TUNDRA AND ITS FAUNA 25

among these we must place the ptarmigan (Lagopus),
of which more than one species occur in the tundra.
In Greenland, in Arctic America and the adjacent

Fic. 2. The Ptarmigan, with nest, young, and eggs, showing the
stony uplands where the bird breeds in Highland Scotland. (From
a specimen in the Royal Scottish Museum.)

regions, in Spitsbergen, in Northern Asia, these birds
occur, sometimes in considerable numbers. In the slow

26 THE TUNDRA AND ITS FAUNA

agony of the last march of de Long and his companions
after the wreck of the Jeannette, it was by the help of
the ptarmigan and reindeer shot that the final cata-
strophe was postponed so long. After Nansen and his
companions arrived in safety at the east coast of
Greenland, their crossing of the inland ice completed,
they whiled away their months of waiting for a ship by
the help of ptarmigan shooting. Sverdrup and his
companions shot many for food during their Arctic
expedition, and similarly almost every expedition which
has landed in tundra regions has noted the presence
and characters of the birds.

The ptarmigan share with the Arctic fox, the stoat
or ermine, and not a few other northern forms the
peculiarity of displaying a seasonal change of colour.
This change doubtless aids these varied animals in
escaping from their enemies, or in stealing unperceived
upon their prey. Like many other tundra animals also,
the ptarmigan is social, being found in flocks, except
at the pairing season. This is no doubt associated with
the fact that food is only obtainable in certain areas,
separated by extensive barren regions. Though in most
favourable tundra regions ptarmigan are to be found
throughout the year, a certain amount of migration
occurs. Thus in Greenland they are commoner in the
south in winter than in summer. Needless to say,
ptarmigan are not peculiar to the tundra, for they
extend far to the south of the region.

The only songster of the tundra is the snow-bunting
(Plectrophenax nivalis), which is very widely distributed,
and is spoken of by most explorers, who welcome its
twittering song in spring. There is no doubt that it is
largely a migrant, nesting in the north, and turning
southwards in the autumn. Occasionally, however,

THE TUNDRA AND ITS FAUNA 27

it appears to winter in the Arctic, notably in Green-
land.

Wherever in the tundra region lemmings and small
birds are to be found, there occurs the snowy owl
(Nyctea nivea), though it does not seem to be common
except where lemmings are abundant, as in Nova
Zembla and parts of the mainland of Northern Asia.
The bird has a general resemblance in coloration to
a ptarmigan in winter plumage, and often lives in the
stony regions frequented by that bird, upon which the
owl preys. It is, however, no exception to the rule that
the carnivorous animals of these barren regions need to
lay the sea under contribution as well as the land, for it
eats fish in addition to birds and small mammals. It
winters in the tundra, and is a very characteristic bird
of the region. Another predatory bird of the Arctic is
the Greenland gerfalcon (Falco candicans), a beautiful
bird, predominantly white in colour, which inhabits
Greenland and Arctic America, and is represented by
an at least closely allied species in Northern Asia. It
preys upon ptarmigan and other birds, and also upon
lemmings, and winters in Greenland. The third pre-
daceous bird of the tundra is the raven (Corvus corax),
which is the scavenger of the region, and will eat
practically anything from land or sea.

As igs to be expected from the unfavourable con-
ditions of life, reptiles and amphibians are absent, and
fresh-water fish are not abundant. Representatives of
the salmon family occur in some lakes, but there is
nothing characteristic.

Among the invertebrates the insects present some
interesting points. Except in Spitsbergen and Nova
Zembla, mosquitoes are enormously numerous, and
form in many regions a terrible plague. It may seem

28 THE TUNDRA AND ITS FAUNA

curious that they should be abundant in this cold
region, but the cause is to be looked for in the un-
developed condition of the drainage system, already
mentioned. The absence of any run-off in winter gives
the tundra, as we have noted, the characters of a semi-
arid region, with the result that the water formed in
summer by the melting of the snow tends to accumu-
late in innumerable lakes and pools. In these the
mosquito larvae swarm, and the natural checks, e.g. in
the shape of small fish who feed on the larvae, must be
insignificant. From the pupal stage the adults emerge
towards the middle of June, and they are then extra-
ordinarily abundant till the middle of August, when
they disappear, after having laid their eggs in the pools.
These pools freeze or dry up with the onset of winter,
but the eggs or larvae remain uninjured, to hatch or
recommence active life with the coming of water in the
spring. It is possible that the life-history is lengthened
by the cold, but the rapidity of reproduction is certainly
not checked, except in the coldest regions. In addition to
mosquitoes a considerable number of other flies occur,
especially about the houses of the Eskimo, &c., breeding
in the refuse thrown out. Indeed, flies predominate
among northern insects, the percentage increasing as
we pass northwards. The reason lies apparently in
the fact that the larval stage is usually short, and
can be run through in the brief period of warmth.
With the exception of the curious parasites known as
Strepsiptera, all the other orders of insects are repre-
sented, the bees and butterflies fulfilling here their
usual role as fertilizers of flowers. The very short
active period of vegetation has a curious effect upon
the development of plant-eating insects. As is well
known, in more temperate countries butterflies usually

THE TUNDRA AND ITS FAUNA 29

lay their eggs in the autumn (except in the case where
there are two annual broods), and these eggs pass the
winter as eggs, or, if the larvae hatch, they winter in
a very undeveloped state. In the spring, activity is
early resumed, and the life-history is rapidly passed
through, so that the butterflies appear in summer. In
Greenland, butterflies do not appear till the middle of
July. If they lay their eggs, and these hatch at once,
the resulting larvae have, in the colder parts, only some
five weeks before winter sets in. Activity does not
recommence till at earliest the first week of the follow-
ing June, and as it cannot be supposed that the animals
can pass through larval and pupal life before the onset
of cold, they must pass a second winter in the larval
state, the butterfly emerging in the following July to
lay its eggs and die. This great lengthening of the life-
history cannot be ascribed wholly to the cold ; we must
suppose that it is also due to the fact that caterpillars
must necessarily wait in spring until their food plants
have made some new growth. This life-history, there-
fore, again emphasizes the baneful effect on animal life
of the scanty development and slow growth in spring of
the plants of the tundra.

In addition to insects, a few earthworms occur in the
warmer parts, and.there are a few land and fresh-water
shell-fish and some spiders and mites. In general, how-
ever, it may be said that among the invertebrates the
insects are the most conspicuous and numerous forms.

As a pendant to this picture of life on the tundra
we may add a brief consideration of the animals of the
shore, which, as we have noticed, have considerable
direct and indirect importance for the land animals.
Man in these regions is practically dependent upon the
sea for food, for fuel, and for clothing, for the land

30 THE TUNDRA AND ITS FAUNA

gives him almost nothing. It is a commonplace of
physiology that in cold climates his need of fat is great,
but in this respect he resembles all the other animals
of the region, all of whom seek fatty food, and store
up in their bodies reserves of this heat-producing sub-
stance. The food from which the larger marine animals
obtain their supplies of fat is almost always, directly or
indirectly,found in the myriads of small crustacea which
haunt polar seas, and whose bodies contain much oil.
Arctic seas are indeed specially characterized by their
wealth of small crustacea and of marine mammals, and
the two facts are closely connected. The crustacea feed
the fish of the polar seas, which, though of relatively
few species, are often extraordinarily abundant as
individuals. The seals and toothed whales feed upon
fish. Small crustacea are also devoured by molluscs,
of which the mud-haunting forms are devoured by the
walrus, and the floating forms by the whalebone whales,
so that the crustacea form the basis of the marine food-
supply. They are in their turn fed by the more minute
forms of life, especially by the minute plants, such as
the diatoms, so abundant in polar seas.

The few survivors of the ill-fated Greely expedition
kept themselves alive by the collection of ‘ shrimps ’
in the shore-water. These ‘ shrimps’ must have been
largely the small crustaceans called amphipods, which
are enormously abundant in polar seas, and the fact
that they were so utilized may be taken as emblematic
of Arctic conditions.

Apart from crustacea, the littoral fauna of the Arctic
is fairly rich. A considerable number of large molluscs
occur, including in the ‘ gapers’ Mya and Saxicava
mussel-like forms, which constitute an important part
of the food of the walrus. Where molluscs abound,

THE TUNDRA AND ITS FAUNA 31

starfish, which prey upon them, are also abundant, and
reach a large size. Where seaweed is abundant, sea-
urchins occur ; according to Nordenskidld, west of Nova
Zembla they are so numerous as almost to cover the
bottom. The mud holds a considerable number of
marine worms, and other less striking invertebrates
also occur in numbers. Here, however, we must note
that if the sea contributes a wealth of food to land
animals, yet the land also gives something to the sea.
The icebergs which detach themselves from the glaciers
carry much rock-waste to the sea, and this forms a fine
mud very favourable to animal life, for very many
shore animals are mud-eaters. The ice-foot and drift-
ice also contribute to the rock débris spread over the
bottom, and the proximity of the glaciers to the shore
in many places must mean that the rock-waste is
carried by the turbid streams of summer direct to the
ocean, instead of being spread over the land, as is for
instance so much of the glacial mud of the Alps. In
other words, the mud which in Central Europe forms
such vast fertile areas as the Plain of Lombardy, is in
Arctic regions carried seawards to help to feed the
swarming life of the sea. It is this fact which explains
the wealth of the Arctic in marine birds and mammals,
and its poverty in land forms of either. The poverty
of the land is made up for by the wealth of the sea.
The characteristic marine mammals of the extreme
north are the walrus and the earless seals, all of which
spend a considerable part of their life on shore, espe-
cially at the breeding season, and the whales, toothed
and whalebone, which never voluntarily come on shore
at all. All have, of course, had terrestrial ancestors,
and we may see one justification of the evolution of
their adaptations to a marine habitat in the fact just

32 THE TUNDRA AND ITS FAUNA

emphasized—the poverty of the land and the relative
wealth of the sea in these northern latitudes. Another
cause of the abundance of the seals and their allies in
Arctic and Antarctic waters is certainly to be sought
in the fact that these animals must come on shore to
breed, and suitable regions for this purpose, offering
security against attack, except in the case of man, the
great enemy of the large mammals, are to be found in
the frigid rather than in the temperate or torrid zones.

Of the Arctic seal-like animals, the walrus (Trichechus
rosmarus), which is circumpolar, is one of the most
interesting forms. As already noted, it is stated to
feed chiefly upon molluscs, but apparently adds fish,
swimming birds, and even seals to these. It seems to
pass a considerable part of its life on shore, or at least
on floating ice-floes. The massive tusks are used in
digging up mussels from the mud, and also in helping
the animal to clamber upon the ice.

The true seals are represented by a number of species,
five or six occurring in Greenland. The presence of
a number of nearly related forms within a limited area
means a very perfect adaptation to the natural con-
ditions. Analogous conditions are presented by the
many kinds of antelope which are found in the open
plains of Africa, and the many kinds of deer in the
forests of Asia.

The true seals are very clumsy on land, but are very
perfectly adapted for swift movement in the sea. They
feed upon fish, to which at times they add shell-fish and
crustaceans. In the sea their great enemy is the killer
whale, but their enemies on land are much more serious,
consisting of the polar bear and man, both of whom
take advantage of the animal’s relative helplessness on
land, and the fact that they must come on shore to

THE TUNDRA AND ITS FAUNA 33

breed, and that the young must spend the first two or
three weeks of their life on shore. Many of the seals
are markedly migratory, going south in winter to avoid
the intense cold of the north. This is especially true of
the Greenland seal (Phoca groenlandica), which seems
incapable of making a breathing-hole in the ice, and
therefore must have relatively open water in its haunts.
On the other hand, the ringed seal (Phoca hispida),
which does make such breathing-holes, is almost non-
migratory, living permanently in the coldest regions.
These two species will thus not compete with each
other, and all the species doubtless show similar
minor adaptations, which fit them for different parts
of the common environment; but it is not always
possible to explain wherein these special adaptations
consist. The large bearded seal (Phoca barbata) is
another form which makes breathing-holes. The very
curious crested seal (Cystophora cristata) is not only
migratory but also pelagic in its habits, avoiding the
neighbourhood of land, and bringing forth its young
on ice-floes. It is a bold species, apparently a swift
swimmer, and eats cuttles as well as fish. The common
seal (Phoca vitulina) also occurs in Arctic waters.

The cetaceans of the north are numerous, and in-
clude some very large animals. Thus among the whale-
bone forms we have the Greenland whale (Balaena
mysticetus), with its enormous head, and its elaborate
series of baleen plates, which filter the minute crus-
taceans and molluscs upon which it feeds from the
sea water. The huge rorquals (Balaenoptera), though
found in practically all seas, often seek the vicinity of
northern coast-lines in spring for breeding purposes.
Among the toothed whales the narwhal (Monodon
monoceros) is a true northern form, and lives largely

1404 c

34 THE TUNDRA AND ITS FAUNA

uponcuttlesand crustaceans. The white whale (Delphin-
apterus leucas) is similarly an Arctic form, remarkable
for its habit of ascending rivers for a considerable
distance. The killer whale (Orca gladiator), on the
other hand, which is a very widely distributed animal,
chiefly occurs in the Arctic in summer.

If we sum up the facts in regard to the distribution
of tundra animals, and of the marine forms which
occur round its seaward margin, we may say briefly
that the tundra fauna is virtually uniform throughout.
The animals, generally speaking, are either circumpolar
at the present time, or have once been circumpolar,
and there is no distinction between eastern and western
faunas. In other words, the fact that the * salt, estrang-
ing sea’ is here more or less ice-bound makes the land
areas virtually continuous, and this, combined with the
marked uniformity of physical conditions, leads to
practical uniformity of distribution, both of plants and
of animals. No such similar uniformity exists in regard
to the,other natural regions of the world. Though the
steppes of Asia show a general resemblance in physical
conditions to the steppe-like regions of North America,
yet their faunas are different. In the same way, the
faunas of the similar equatorial forests of South America
and Africa are markedly dissimilar, and there is almost
no resemblance between the desert fauna of Australia
and those of Africa or of America. No resemblance,
that is, when the actual affinities of the animals are
considered, but necessarily desert animals or forest
animals, whatever their affinities, share certain adapta-
tions to forest or desert life.

RerErences. Accounts of tundra animals are to be found in almost

all books dealing with the Arctic regions. The following may be men-
tioned as giving a considerable amount of detail: Manual of the Natural

THE TUNDRA AND ITS FAUNA 35

History, Geology, and Physics of Greenland and the Neighbouring Regions,
edited by Professor T. Rupert Jones (London, 1875); The Voyage of the
Vega, by A. E. Nordenskiéld (English Translation, London, 1881) ;
Grinland-Expedition der Gesellschaft fir Erdkunde zu Berlin, unter
Leitung von Erich von Drygalski (Berlin, 1897). The second volume of
the last named gives a very full account of the fauna and flora of Green-
land by Dr. Ernst Vanhéffen. The animals of polar regions generally
are discussed in Dr. Bruce’s Polar Exploration (London, 1911). In
addition the accounts of their explorations published by Nansen, Sver-
drup, the Duke of Orleans, the books dealing with the Jeannette ex-
pedition, &c., may be consulted. Kobelt’s Die Verbreitung der Tierwelt
(Gemissigte Zone), published at Leipzig in 1902, discusses tundra
animals in general, and gives some characteristic views of tundra life.
See also Seebohm’s Siberia in Asia (1882), and the same author’s Address
to Section E (Geography) at the Nottingham Meeting of the British
Association, 1893.

Q
ro

CHAPTER II

THE TAIGA, OR CONIFEROUS FOREST, AND
ITS FAUNA

STRIKINGLY different from the fauna of the tundra
is that of the temperate forest, which forms a wide
belt in the northern hemisphere, lying just south of
the tundra. Here the climate is more favourable,
especially in the southern portion, and the flora far
more luxuriant. Nevertheless here, as in the tundra,
the ebb and flow of plant life is caused by the seasons,
that is, by the varying amounts of heat received. In
subtropical countries, with their winter rainfall, it. is
water, not heat, which determines vegetative activity.
In the tropics, except where there is a well-marked dry
season, vegetative life is practically continuous, but in
the temperate forest zone, no less than in the drier
steppes, it is the alternation of summer and winter
which is of importance. As all animal life in these
regions ultimately. depends upon land plants, this means
that there are periodic alternations of plenty and »
scarcity. In the forest zone, however, this is less
marked than in the steppe zone. In the coniferous
forests the trees for the most part keep their leaves all
the year round. However deep the snow, the twigs
and branches of the trees are exposed, and these twigs
and branches very often carry nuts, berries, or seeds,
which persist throughout the winter, and thus prolong
the summer abundance into the time of scarcity.
Further, the forest gives shelter, and thus diminishes

THE TAIGA AND ITS FAUNA 37

the rigour of the winter climate. The thick branches
may shelter the ground from snow, the covering of
dead leaves may protect it from frost; hollow trees
and dense bushes offer protection to many small
animals. In short, while the forest does not offer the
rich pasturage of the steppe in summer, it offers greater
advantages in winter, and especially greater reserves
of food.

One result of this is that on the whole the habit of
migration is less marked in the animals of the temperate
forest than in steppe ones. It does occur, but, on the
other hand, both tundra and steppe animals may seek
the forest in winter, and the greater uniformity of con-
ditions in the forest prevents migration occurring on
the same scale as in steppe and tundra, where areas
of abundance even in the summer season alternate.
with others where famine reigns.

As a consequence of the fact that the forest offers
a refuge to animals from other areas, we find that its
denizens may be divided into two groups. We have
forms like the squirrel, the dormouse, the lynx, the
woodpecker, the tree-creeper, the tree-frog, and so
forth, which are structurally adapted for forest life,
and are not at home elsewhere, and we also find animals,
like some kinds of deer and wolf and fox, which show
no very special adaptations to forest life, but which
visit the forest for shelter or for food.

We have already noted that the northern limit of
the forest is determined by adequate summer warmth.
Its southern limit depends upon much more complicated
factors, especially upon the rainfall, and the absence of
drying winds in winter, when the coldness of the soil
prevents absorption. A small rainfall, limited to the
summer months, favours grass rather than trees, and

38 THE TAIGA, OR CONIFEROUS

their slow rate of transpiration makes conifers more
tolerant than broad-leaved trees of unfavourable con-
ditions. The consequence is that in Asia we find
a band of predominantly coniferous forest separating
the tundra from the steppes of the central region, this
forest, which is markedly uniform in character, being
called the taiga. Its important trees are the Siberian
larch and the Siberian fir ; the Cembra pine, important
for its edible seeds, eagerly relished by many animals ;
with many minor forms, such as juniper, bird cherry, °
mountain ash, &¢., important in that they bear edible
fruits. In addition the taiga resembles the tundra in
its great wealth of berry-bearing bushes, such as
cranberries, crowberries, whortleberries, &c., especially
abundant in clearings, and in areas devastated by
forest fires.

Lianes and creepers are typically absent, and the
taiga as a whole is much less dense than the tropical
forest. The result is that we have no animals of purely
arboreal habit, such as occur in the equatorial forests,
and on the whole arboreal adaptations are few. Very
many of the inhabitants of the taiga possess the power
of climbing without showing any great modification of
structure.

In North America, especially in the eastern part of
Canada, a belt of forest of very similar characters occurs.
The species of conifers are different, but the general
aspect of the forest is the same, and its inhabitants
show a general resemblance to those in the Old World.
Under the name of taiga animals, therefore, we shall
describe the inhabitants of the coniferous forests of
both the eastern and western hemispheres.

The important mammals of the taiga belong to three
orders ; they are ungulates, or rodents, or carnivores.

FOREST, AND ITS FAUNA 39

There are in addition a few bats and a few insectivores,
but the predominant forms belong to the orders named.
We shall begin with the ungulates, as including the
largest and most obvious of the inhabitants of the
forest.

With the exception of the wild boar (Sus scrofa),

Fig. 3. The Wapiti of the American forests, often miscalled Rocky
Mountain ‘elk’. (Photo by the Biological Survey, U.S.A.)

which just penetrates the Asiatic taiga from the south,
and excluding from consideration such forms as sheep
and goats, really mountain animals, we may say that
the ungulates of the taiga consist of deer. Deer are
typically forest-haunting animals, though some of them
feed on the outskirts of the forests, and a few species
inhabit open plains, especially in South America, where
large hoofed mammals are few. Except in South
America, the place of deer on grassy plains is taken by

40 THE TAIGA, OR CONIFEROUS

antelopes, especially abundant in Africa, where deer are
altogether absent. High up on mountains also, their
place tends to be taken by sheep, goats, and. allied
forms. The Eurasiatic continent, which is rich in
forests, is rich also in species of deer, but many of
these are confined to tropical and subtropical regions,
leaving relatively few to occur in the taiga proper.
So far as these forms go there is a close resemblance
between the eastern and western halves of the
continent.

The fact that deer are characteristic animals in the
taiga is in itself sufficient to enable us to draw certain
conclusions in regard to the nature of the region. When
ungulates occur in tropical forests they are mostly
small animals, with narrow compressed bodies, which
allow them to glide through the brushwood. Deer
are not only big animals, but the broad, spreading
antlers of the male would make rapid movement
through a dense wood impossible for him. Some
authorities regard these antlers as evidence that deer
have only recently taken to wooded regions; their
presence at least excludes the animals from very
dense thickets.

In both hemispheres there are taiga varieties of the
reindeer, and both have elk or moose (Alces machlis).
In the Asiatic taiga, the red deer (Cervus elaphus)
occurs, with closely related forms of similar habits,
such as the maral stag (C. maral). Though these forms
do not occur in Canada, the wapiti (C. canadensis) is
a nearly related species. America has in addition the
Virginian deer (Cariacus virginianus), belonging to
&@ group numerously represented further south, and
distinguished by the great development and curious
structure of the antlers. In the Asiatic taiga the

The Elk or Moose of Alaska.
(Photo by the Biological Survey, U.S.A.)

Fia. 4.

ries
—
*

(Photo by the Biological Survey, U.S.A.)

an Deer.

irgini

Domesticated V:

Fic. 5.

FOREST, AND ITS FAUNA 41

small roedeer: (Capreolus caprea) also occurs, and the
musk-deer (Moschus) is found occasionally, though its
true home lies further south. The last is an aberrant
form, almost hare-like in its habits, and much per-
secuted for the sake of the scent gland of the male
(see p. 85).

Most of the animals above mentioned are at least
largely forest-feeders. That is, twigs and leaves both
of coniferous and deciduous trees, berries, nuts, &c.,
bulk largely in their diet, and grass is relatively less
important. The most northerly forms of those men-
tioned are the woodland reindeer and the elk, both
very characteristic of the taiga. In North America the
woodland reindeer is larger than the tundra form,
perhaps because it is better fed, and chiefly haunts
marshy ground. Within the forest, as outside of it, it
depends largely upon mosses and lichens. In winter it
leaves the swamps for the dense forests on the higher
ground.

In the Asiatic taiga some interesting observations
on woodland reindeer have recently been made by the-
members of Mr. Carruthers’ expedition to Mongolia.
Here, on the Sayansk divide, the members of the
expedition found traces of a forest-haunting form
which spends the summer in the rhododendron scrub,
near the upper limit of the forest. In the winter the
animals seek the open hill-tops, which are swept bare
of snow by the wind, and there feed upon ‘ reindeer
moss ’.

The elk in both the Old and New Worlds is a true
forest animal, compelled to feed upon trees and bushes
by the fact that its long legs and short neck prevent it
from grazing on the ground, except where the grass is
very long.

42 THE TAIGA, OR CONIFEROUS

The red deer seem to feed chiefly on the outskirts of
the forest or in the open glades, but the American
wapiti eats a considerable amount of leaves and twigs.
The Virginian deer also eats a large amount of forest
produce, sharing with many other animals, both birds
and mammals, a great fondness for beech mast, which
it obtains in winter by pawing away the snow. It
extends into Canada, but its true home is rather further
south than the taiga proper.

The most important rodents of the taiga are the
various kinds of squirrels, such as the true squirrels
(Sciurus), the striped squirrels or chipmunks (Tamias),
and the flying squirrels (Sciuropterus); the beaver
(Castor); the Polar hare, which extends into the
forest region, though it is not common there. The
American taiga also lodges the Canadian porcupine,
an arboreal animal, the musquash (Fiber), a swamp
animal, and the American variable hare in addition to
the Polar hare.

The squirrels form a very interesting group, showing
progressive adaptation to arboreal life. The Siberian
striped squirrel or chipmunk (7’. asiaticus) is an example
of a ground form which occurs in both North America
and Asia, though North America has other species in
addition. The chipmunks feed on nuts and seeds, of
which they store large quantities for the winter, and,
like the true squirrels, they hibernate. They differ,
however, in living mostly on the ground, where they
often excavate holes. They can climb trees, but are
not nearly so agile as the true squirrels. The next
stage is represented by the American red squirrel or
chickaree (S. hudsonianus), which is a graceful and
agile climber, but spends a good part of its time upon
the ground, where it sometimes excavates holes. Our

FOREST, AND ITS FAUNA 43

common British squirrel (8. vulgaris), widely distributed
in the more northerly parts of Asia, as is well known,
rarely descends to the ground, and has great powers of
leaping. It shows interesting colour variations, tend-
ing to be darker in colour in the colder parts of its
range in summer, and turning white in winter where
the climate is severe. It will eat almost any forest
product, and hibernates during winter. The final stage
in the mastery over the forest is represented by the
flying squirrels, which are furnished with a parachute
by means of which they can take long leaps from one
tree to another, and thus obviate the risks due to
a descent to terra firma. Though chiefly inhabitants of
the tropical forests of Asia, one species occurs in tem-
perate Asia and one in temperate North America.

The beaver, which haunts the streams flowing through
wooded regions, was once widely distributed over the
cooler parts of the northern hemisphere. The American
species, if not identical with the Eurasiatic one, is at
least very nearly allied. Beavers feed upon bark and
twigs, and also upon the roots, &c., of such plants as
water-lilies.

While the porcupine of the warmer parts of Asia
inhabits rocky country, the porcupine of Canada is
a purely arboreal animal in habits, though it does not
possess the same adaptations to this mode of life as
are exhibited by its southern allies, the tree-porcupines
of South America, Mexico, &c. The Canadian form
(Hrethizon dorsatus) follows the taiga to its northern
limit, and also follows the western coniferous forest to
its southern limit. It spends the greater part of its
time in the trees, though the nest is usually made on the
ground among stones. The food consists of leaves,
bark, twigs, beech mast, and other nuts, and the

44 THE TAIGA, OR CONIFEROUS

animals are stated to defoliate completely the trees
they attack.

As regards the carnivores, the special features of the
northern forests are the relative poverty in true cats,
so abundant in tropical forests generally, and the
wealth of arboreal animals of the weasel family, which
are most abundant in temperate latitudes, and take
there the place of the true cats and civets elsewhere.
The bears also are characteristic of temperate forests,
though not confined to them.

There is a close resemblance between the carnivores
of the American and Siberian taigas. Thus in both are
found lynxes, the common wolf, the common fox, bears,
martens, the glutton, and among less purely forest
forms, weasels, stoats, minks, and badgers. In Siberia,
especially in the south, the tiger is sometimes found,
while in the Canadian taiga the puma (Felis concolor)
does not occur, although it is common in the coniferous
forest of the Rocky Mountains. The wild cat (Felis
catus) of the wooded parts of Europe does not occur in
the Siberian forest, while it is altogether absent from
America. An animal found in the Canadian forests,
but with no Old World representative, is the skunk
(Mephitis mephitica), a member of a group charac-
teristic of South America, and the only one which
ranges far north.

The lynxes are represented by the common lynx
(Felis lynz) in Siberia, and by a closely related species
(F. canadensis) in the woods of Canada. They are
expert climbers, and are true forest animals, but the
Canadian form is stated to feed chiefly on hares and
birds of the grouse family, so that it must find its food
largely on the ground. The common wolf is not
specially fitted for forest life; it lives either in the

FOREST, AND ITS FAUNA 45

woods or in the open, wherever food can be easily
obtained. The fox with its burrowing habits is more
of a forest animal, but only because in the woods it
finds the necessary shelter and abundant food. It has
no special adaptations to life there.

In addition to wolf and fox Siberia has another
member of the genus Canis, in the form of the Siberian
wild dog (C. alpinus), which apparently haunts the
high Alpine forests, as well as open country.

The brown bear (Ursus arctos) occurs in the Siberian
taiga, as in the woods of the temperate region of the
Eurasian continent generally. It can climb, though it
is not so agile as some of its relatives, and, like all the
bears, it takes a large amount of vegetable food. It is
stated to be particularly fond of the edible seeds of the
Cembra pine, and climbs the pines in search of them, as
well as seeking to rob the peasants’ stores. Like its
relatives also it is very fond of honey. In the taiga of
North America it is replaced by the black bear (Ursus
americanus), of similar habits. The grizzly (U. horri-
bilis) is only found in the western forest. Both the
black and brown bears eat fish and other aquatic
animals in addition to warm-blooded forms, nor do they
disdain insects.

In the Siberian forests two kinds of martens occur,
the pine-marten (Mustela martes) and the more valuable
sable (M. zibellina), with a more northerly range and
a more valuable coat. Both are thoroughly arboreal,
though both descend to the ground upon occasion.
Closely related is the American marten (M. americana),
which is of similar habits, while the fisher marten
(M. pennanti), in spite of its name, is also largely
arboreal. It occurs throughout the greater part of
North America, haunting mountain regions in the

46 THE TAIGA, OR CONIFEROUS

warmer parts of its range. The polecats, of which one
(IM. eversmannia) occurs in Siberia, the common weasel
(M. vulgaris), found both in Northern Asia and in
Northern America, and the stoat or ermine (IM. erminea)
as already indicated, are not definitely forest animals,
though they are found there as well as elsewhere. The
mink (M. vison) of North America, and the Siberian
form (M. siberica) haunt streams and lakes, and are
forest animals to the extent that localities suitable to
their habits often occur within forest areas.

On the other hand, the wolverene (Gulo luscus),
a fierce and voracious carnivore, is chiefly found within
the forest, where it preys upon all manner of small
mammals and birds, and even attacks weakly or
disabled deer.

The common badger (Meles tawus) is found in the
Siberian forests, and, though a burrowing form, occurs
chiefly in woods. On the other hand, the American
form (Taxidea americana) is chiefly a prairie mammal,
and feeds largely upon the prairie marmots (see p. 64).

Passing now to the birds, we find that the forest is
rich in specially adapted forms, and that in addition
there are large numbers of migrants who spend part of
their time there. The forest gives relative security
during the dangerous period when brooding is taking
place and the young are being reared. The absence or
paucity of snakes in the northern forest also, and the
complete absence of monkeys as well as of the many
carnivorous or egg-eating animals which haunt tropical
forests, increase the value of the northern forest from
the birds’ standpoint, and help to account for the
number of species found there in summer time. Among
forms which are especially abundant in northern
forests, or especially adapted to life there, we may

FOREST, AND ITS FAUNA 47

note the grouse and their allies ; the grosbeaks and their
relatives the crossbills ; the woodpeckers and wryneck ;
the nutcrackers, jays, and their allies.

Grouse belong to the same family as the ptarmigan,
and are especially characteristic of northern latitudes.
The abundance of berry-bearing bushes, of which we
have already spoken, both in the tundra and in the
taiga, is an important factor in their distribution, for
these berries constitute a considerable part of their
food. In the Asiatic taiga we find the black grouse
(Lyrurus tetrix), the larger capercaillie (T'etrao urogallus),
and the hazel grouse (T'etrastes bonasia), while in the
Canadian forest is found the Canadian grouse (Cana-
chites canadensis), often called a partridge, and also the
ruffed grouse (Bonasa umbellus).

The woodpeckers are birds showing remarkable
adaptations to an arboreal habitat. The feet are well
fitted for climbing, and the most highly specialized
forms have stiffened tails, which aid in the climbing
process. The beak is powerful and chisel-shaped,
enabling the birds to excavate holes in wood, and also
to lever off bark, while the long, viscid, worm-like
tongue enables them to capture the insects thus
exposed. Woodpeckers are very widely distributed,
a few only ranging into the northern woods. Among
these are the green woodpeckers (Gecinus), found in
the temperate regions of the Old World, and the
spotted woodpeckers (Dendrocopus), of both Old and
New Worlds. Related is the curious wryneck (Iynz
torquilla), found in summer in Northern Europe and
Asia, which is without the stiff tail and climbing habits
of the woodpeckers.

We have spoken of the edible seeds of the Cembra
pine, and of the appreciation in which they are held

48 THE TAIGA, OR CONIFEROUS

by many inhabitants of the Siberian taiga. Though
the seeds of other conifers have not the same wealth

Fic. 6. The Three-toed Woodpecker of North America. Note the
characteristic attitude, and especially the position of the tail. (Photo
by the Biological Survey, U.S.A.)

of food products, yet they also are edible to many
animals, and forma a much sought after food supply. In

FOREST, AND ITS FAUNA 49

both the Siberian and Canadian taiga they form a very
large part of the natural food of the pine grosbeak
(Pinicola enucleator), a gorgeous bird related to the
bullfinch, and of the crossbill (Louia curvirostra), also
with bright-coloured plumage. Pine cones, it will be
remembered, ripen slowly, and during the ripening
process the seeds are carefully protected by the closing
of the scales, so that their extrication is a matter of
difficulty. The forester is constrained to apply heat to
obtain the seeds without damage, but the crossbills
have the beak converted into an instrument by means
of which the seeds are readily prized out. More slashing
are the methods of the nutcracker (Nucifraga caryo-
catactes), which in the Alps, as in Siberia, may be seen
breaking the cones of the Cembra pine to pieces with
its powerful bill. The western coniferous wood of
North America is inhabited by another species. Re-
lated to the nutcrackers are the magpies and jays, the
latter represented in the eastern taiga by the Siberian
jay (Perisoreus infaustus), with soft fluffy plumage of
a reddish colour.

Reptiles and amphibians are few in number in the
northern forest, the former especially. Among am-
phibia, frogs and toads haunt the pools and swamps,
an important form being the large bull-frog (Rana
catesbiana) of North America, which is exceedingly
voracious, devouring fish, young water-birds, &c., and
derives its common name from its noisy croak, which is
compared to the roaring of a bull. The tree-frogs
(Hylidae), with their usually greenish colour and the
curious adhesive pads at the ends of the toes, by means
of which they keep their foothold, have a much wider
distribution in America than in the Old World. In
the Eurasiatic continent the common tree-frog (Hyla

1404 D

50 THE TAIGA, OR CONIFEROUS

arborea) of Europe just reaches the south of the taiga,
but the tree-frogs of North America have a more
northerly extension.

In insects the northern forests are rich, for the
perennial vegetation feeds many herbivirous forms,
and many carnivorous or parasitic forms prey upon
these. Many of the beetles especially are important
pests from the point of view of the forester.

As examples of wood-eating forms we may mention
the wood-wasps (Sirex), of which a large species (8S.
gigas) often appears in summer-houses or other wooden
erections in Britain, where it has been introduced with
pinewood. The adult female has a boring apparatus,
by means of which she bores a hole in trunks of pine
trees in order to lay her egg at the bottom of the
excavation. The wood-wasps are preyed upon by
ichneumon flies, whose delicate senses enable them to
find the burrows of the wasps, when they insert their
long ovipositors and lay their eggs close to the wasp’s
egg, so that the resultant larvae may prey upon the
wasp larva. There is here a singularly delicate adjust-
ment between host and parasite, comparable to the
relation which exists between the length of the pro-
boscis of certain butterflies and the length of the tubes
of the flowers which they visit.

Another form which feeds in the pine woods is the
pine saw-fly (Lophyrus pint), whose larvae feed upon
the needles, sometimes in countless numbers.

Wild bees also occur in the woods, the forms which
store up honey extending much further north in the
Old World than in the New. In Asia wild honey bees
reach the Arctic circle, while in North America, accord-
ing to Marshall, they scarcely pass latitude 50° N. The
honey is greatly prized by bears, who rob the nests

FOREST, AND ITS FAUNA 51

without regard to the stings of the bees, from which
their thick coats doubtless protect them.

REFERENCES. Kobelt’s book, already mentioned, gives a full account
of the taiga animals, and descriptions will also be found in Brehm’s
From North Pole to Equator (in translation, London, 1896). As the
inhabitants of the taiga are mostly those characteristic of the temperate
regions of the Northern Hemisphere, full accounts of them will be found
in the books mentioned at the end of chapter x. Most descriptions of
journeys through Siberia also give some account of the taiga animals.

CHAPTER III

-STEPPE FAUNAS AND THE TEMPERATE
STEPPES OF ASIA AND NORTH AMERICA

To the south of the coniferous forest of Asia lies
a belt of varying width, characterized by the absence
or scarcity of trees, and by the periodical growth of-
grasses and other herbaceous plants. This belt of
steppe land is continually merging into desert, the cold
desert of the high uplands, or the warm desert of the
rainless regions to the south. Similarly, the forest sends
long feelers into it, wherever the banks of a water-course,
or some favouring condition of soil or local variation of
climate, make this possible. As Central Asia generally
is a country of high mean elevation, and as it is remote
from the sea, the climate is continental in character,
very hot in summer and very cold in winter. The
region is also continually swept by very severe storms,
injurious alike to plant and animal life. Further,
though the rainfall is always slight, and chiefly con-
fined to the summer period, there is some reason to
believe that it is subject to great variation, perhaps
cyclical in character.

This very brief description may be said to sum up
the main features of regions of temperate steppes, and
the peculiar conditions cause them to be inhabited by
special types of animals. Some of the conditions which
we noted in the tundra are here repeated, for indeed
the tundra is but a special type of steppe. Thus in
both we have a seasonal.and highly local abundance of

STEPPE FAUNAS OF ASIA AND AMERICA 53

food, and also a seasonal and local scarcity. If the
rains of early summer are abundant in the steppe there
is an extraordinarily rapid awakening into life on the
part of the steppe plants, resulting in a great abun-
dance of food for the herbivores of the region. If the
rains fail, scarcity reigns, and it comes in any case as
soon as the favourable season is past ; that is, alike in
the height of summer and in the depth of winter. But
local conditions produce minor variations in climate,
altitude being one of the most important of these con-
ditions. For this reason there are frequent alternations
of famine and plenty, alike in time and space. Like
the tundra animals then, the steppe animals tend to be
social, to. move about in flocks from one region of
abundance to another. But as the wealth of the steppe
is greater than that of the tundra, they are far more
numerous, and far more diverse. Like the tundra
animals as a whole, they attempt to escape unfavour-
able conditions by migration, and as these migrations
must be rapid, to prevent death by starvation occurring
between one region of pasturage and the next, the
steppe animals are mostly swift, with special adapta-
tions to ensure rapidity of movement.

The herbivores, always, as we have seen, necessarily
the majority, are far more exposed in the open steppe
than in the forest, and therefore they are either bur-
rowers, or have singularly keen senses, enabling them
to perceive danger from afar. The social instinct also
is here of great aid. In some cases, as with the wild
asses, the strength of the male enables him to protect
the females and young from attack. As the safety of
the species depends in this case upon the strength of
the male, we find that he must win and keep his post
as leader and defender by his strength, and that he is

54 STEPPE FAUNAS AND TEMPERATE

deposed by a younger male so soon as his powers begin
to fail.

Again, the social instinct is an aid in another way.
The more helpless forms, such as the rodents, usually
appoint sentinels when they feed, these sentinels giving
warning of the approach of danger in time for the
party to seek safety in flight or underground. In the
forest, where food is more uniformly distributed and
the natural shelter greater, social animals are less
frequent.

Another marked feature of steppe animals is their
great fertility, seen alike in the steppe rodents and in
insects like the locusts. This is associated with the
risks of the natural habitat—risks of drought and con-
sequent lack of food, risks of storm, risks associated
with winter cold and summer heat. In Central Asia
the dreaded ‘buran’ or hurricane may practically
exterminate all life within a given area, except such
animals as can find a refuge underground. As such
storms are frequent, great natural fertility is necessary
to repeople the devastated regions. Similarly, a season
of deficient rainfall, or a series of such seasons, must
kill out large numbers of animals.

We have already seen that this fertility of steppe
‘ animals is associated with strong migratory instincts.
Now no natural barriers separate steppe regions from
the neighbouring areas of forest, of desert, of semi-
desert, and soforth. We find then, asa special character
of steppe animals, the fact that they tend periodically
to overrun the means of subsistence within their own
region, and therefore to flow out into the neighbouring
regions. The direction of the winds and the position
of Europe on the western border of a great continent
give it a moist climate, and led to its being forest-clad

STEPPES OF ASIA AND AMERICA 55

almost throughout till man interfered, though the steppe
region of Asia is continued into it through Southern
Russia and part of Hungary. We find then that since
the close of the glacial period Europe has been con-
tinually liable to incursions of steppe animals from
Asia, some of which, like the hamster, have kept their
hold, while others, like the saiga antelope, rapidly
died out.

Another character of the animals of the temperate
steppe is the prevalence of hibernation among them.
The tundra is too cold for hibernation to be practised
to any great extent, and the shelter and presence of
food even in winter in the forest makes it relatively
infrequent there. But the animals of the steppe can
only find in burrows and in sleep protection against
want of food and extremes of temperature, so that here
both aestivation and hibernation may occur. The
squirrels in the fir woods wake whenever the temperature
rises in winter, just as the fir-tree itself becomes active
whenever the temperature permits. But the plants of
the steppe disappear beneath ground so soon as their
brief period of activity is over, and their quiescence is
reflected in that of many of the steppe animals.

Large areas of land almost devoid of trees, and
clothed, at least at times, with grasses and herbs, occur
in many other.temperate regions besides Asia and
South-eastern Europe. A great area of land of this
type extends from the border of the Canadian forest
southwards through the United States in the region
west of the Mississippi river, nearly to the Gulf of
Mexico. Similar areas occur in the Argentine and in
Patagonia in South America, in South Africa, and in
the south-eastern part of Australia. By extension the
term steppe can be applied to all these regions, and

56 STEPPE FAUNAS AND TEMPERATE

there are other large areas in Africa and in the north of
South America where open park-like country occurs,
with trees scattered among the grass, and such land has
some of the characters of a steppe. Again, all the great
deserts of the world are fringed by semi-arid areas,
displaying the general characters of a steppe.

The great interest of the steppe areas is that, espe-
cially in Asia and Africa, they are the natural home of
very many of the great ungulates, the most highly
specialized of which are steppe animals. In North
America the steppe lands, within the human period
at least, have been inhabited only by few species of.
wild ungulates, but of these few the bison was repre-
sented by countless numbers of individuals until it
was exterminated by the white man. In South America,
within the human period, but prior to the immigration
of the white, large ungulates were almost absent. No
horse, no relative of cattle or sheep or antelope cropped
the herbage of the great plains, but their place in
nature was taken by enormous numbers of rodents,
which reached here a size not attained elsewhere.
Again, in Australia no ungulate whatever occurred,
and the natural pasture was utilized by marsupials or
pouched animals, of which the most important grass-
eating form is the kangaroo.

That South America has few native ungulates and
Australia none are two of the most interesting facts in
geographical distribution, but the fact that ungulates
introduged by the white man into both countries have.
flourished apace makes it unnecessary for us to suppose
that any natural obstacle to their presence there
existed. We may then say generally that the steppe
regions of the world are the regions which form the
natural home of the ungulates most valuable to man.

STEPPES OF ASIA AND AMERICA 57

It is the steppes of the more remote parts of the world
which now support the vast herds of sheep, horses, and
cattle necessary for the wants of civilized communities.
Further, as steppe regions do not require to be cleared
of trees, and, where the rainfall is sufficient, often pro-
duce abundant crops, we find that it is in the steppe
regions that the cultivation of cereals is being most
vigorously pushed at the present time. The regions of
dense population are mostly regions of abundant rain,
and therefore were once forest regions. Formerly these
watered lands were liable to be periodically flooded by
the overflow of the fertility of the steppes. This great
tide is now being gradually regulated, and the steady
stream of wheat, of wool, of meat, and so forth, which ,
pours into the civilized world, is the biological equiva-
lent of the flocks of rodents and the flights of locusts
which the steppe sent out in earlier days, and of the
hordes of wandering nomads which came later.

To this general account we may add a few words on
the special conditions, climatic and other, found in the
steppes of temperate Asia, the region to which the
following description specially refers. Steppe condi-
tions reign over that great area of Central Asia which
is practically ringed by mountain chains, as well as to
the west of it, but the fauna throughout this area is
not uniform. The mountain chains themselves have
their own characteristic animals, and the plateau of
Tibet, with special conditions of climate, has a peculiar
fauna of its own. Excluding Tibet and the mountain
chains then, we have a great steppe and desert area
which extends from the eastern base of the Pamirs to
the Khingan Mountains, and is separated from the
forest region to the north by the great area of elevated
ground in which the Siberian rivers arise, and from the

58 STEPPE FAUNAS AND TEMPERATE

wooded regions of China to the south-east by mountain
chains. Through the Dzungarian Gate this area, to
which the general name of Desert of Gobi or Shamo
may be given, is continuous with the extensive band of
steppes which runs through Russian Turkestan to the
Aral and Caspian seas, is continued through Southern
Russia in Europe into Hungary, and is bounded to the
south-west by the plateau of Persia. But uniform con-
ditions do not reign throughout this great area, which
shows many alternations of fertility and aridity. Thus
the snows of the Pamirs water much of Russian Turke-
stan through the Amu Daria and the Syr Daria, and
this westerly region as a whole has a much milder
winter climate than the easterly region, permitting of
the extension into it of animals really peculiar to the
southern steppe and desert regions, those characterized
by the absence of severe winter cold.

Climatically, as we have seen, the general features of
the steppes of temperate Asia are the great range of
temperature, the hot summers and the bitterly cold
winters, and the small precipitation, which tends to
occur in late spring or early summer, giving a dry
winter and a dry summer and autumn, which makes
the growth of trees virtually impossible. The frequency
of dry east winds in winter is an important feature in
checking the growth of perennial plants, except such as
die down to the ground. As indications of the nature
of the climate, we may note that at Orenburg the
mean January temperature is 4° F., and the mean July
70° F., while the mean annual precipitation is only 17’,
with a June maximum, but no absolutely dry month.
Further to the east much lower winter temperatures -
occur ; thus Sven Hedin found temperatures of —17° F.
in the Tarim basin,’and the river here is frozen for three

STEPPES OF ASIA AND AMERICA 59

and a half months every year. Even the Amu Daria,
in a western region of relatively mild winters, is frozen
for about a month every year. The steppe animals are
thus subjected to very unfavourable natural conditions,
but the high summer temperature, especially during
the rainy season, means a great intensity of life for
a short period.

In giving a brief account of steppe animals, we shall
limit ourselves chiefly to the steppes of Central Asia
just described, and, as before, will begin with the
ungulates, here especially important.

Perhaps the most characteristic of the steppe ungu-
lates is the saiga antelope (Saiga tartarica), an ungainly
animal with rather short legs, the male having a peculiar
swelling on the face, which makes it appear hook-nosed.
The animal occurs in thousands in the Kirghiz steppes,
and the fact that its yellowish coat becomes almost
white in winter suggests the severity of the climate in
its home. Formerly it had a much wider distribution
to the west, its remains having been found even in
Southern England, as well as in Belgium and Southern
France. In the historic period it has been gradually
retreating eastwards, thus losing the ground which its
ancestors conquered in the ages when steppe con-
ditions were more widely spread than at present. The
antelopes are also represented in the steppes by two
kinds of gazelles, the so-called Persian gazelle (G. sub-
gutturosa), which extends to the Gobi desert, and the
Mongolian gazelle (G. gutturosa), a somewhat larger
animal; but, generally speaking, the gazelles are more
characteristic of the warmer deserts to the south.

No less than three kinds of horse-like animals haunt
the Asiatic steppe—the tarpan or wild horse (Equus
caballus), Prejevalski’s horse (EZ. prejevalskit), and the

60 STEPPE FAUNAS AND TEMPERATE

kiang or wild ass (Z. hemionus). The tarpan is now
limited to the regions of the steppe most remote from
human influence. Like its allies it goes about in small
troops under the leadership of an old male, and has
extraordinarily acute senses and great swiftness. In
winter its coat is so thick as to form a kind of fur,
and it obtains food by scraping away the snow from
the ground. It is almost uniform in coloration, and of
a dun colour. Prejevalski’s horse is of a similar colour
above, but is almost white underneath. Such a dun
tint is common in steppe and desert animals, and is
doubtless protective. In the kiang there is a dark
stripe down the back, and traces of striping on the
limbs, but the body is of a uniform darkish colour,
instead of being striped as in the African zebras. The
kiang (see Fig. 16) inhabits the higher parts of the
steppes, and occurs at great heights in Tibet.

Still another group of ungulates is represented in the
steppes by the Bactrian or two-humped camel (Camelus
bactrianus), which with its long hair, moderately short
legs, and hard feet is as well fitted for the hilly and
rocky regions of the steppes, as is the one-humped
camel for the sandy deserts to the south. The animal
feeds largely upon the bitter saline plants of the steppes,
and will drink saline and brackish water, so that it is
a typical inhabitant of the salt steppe. The two humps
allow for the storage of fat when the animal is well fed,
and thus permit it to withstand periods of semi-starva-
tion. Its appetite is, moreover, remarkably catholic,
for, according to Prejevalski, it will devour all sorts
of animal matter if vegetable food fails.

Taking antelopes, horses, and camel as representa-
tives of the ungulates of the steppes, we may pass on
to the consideration of the rodents, which are numerous

STEPPES OF ASIA AND AMERICA 61

both as to individuals and as to species, and often show
some curious adaptations to steppe life.

A point of very considerable interest in regard to
the steppe rodents is that there is much general resem-
blance between those inhabiting Asia and those of
North America. This is the more remarkable in that
there is no resemblance in regard to the ungulates of

r
Fie. 7. Transport Camels in the Australian Desert.

the two regions. The prairies of North America have
no antelope, no camel, no wild horse nor ass. Their
chief ungulate was formerly the bison (Bos americanus),
while the bison of the temperate regions of the Old
World was a forest-dwelling animal, found in Europe,
but not in the steppes of Asia. The other important
ungulate of the North American prairies is the prong-
buck (Anitilocapra americana), an animal related to the
antelopes, but differing markedly in the fact that the

62 STEPPE FAUNAS AND TEMPERATE

horns are shed annually and are branched. The con-
trast between the living, or recently exterminated,
ungulates of the Old and New Worlds is the more
remarkable when we find that America has many fossil
horses, though no living ones.

In proof of the statement just made as to the resem-
blance between the steppe rodents in the Old and New
Worlds, we may note that marmots occur in both
regions, that the susliks of Asia are represented by the
forms called gophers in America, which belong to the
same genus, and that the jump‘ng mouse of the American
prairies takes the place of the jerboas and their allies
in the Old World. As the adaptation to steppe life is
similar, we shall refer chiefly in the following descrip-
tion to the rodents of the Asiatic steppes.

The susliks (Spermophilus) are related to the squirrels,
but the common steppe form (S. citillus) differs in the
very short tail and the minute ears, both adaptations
to the terrestrial and burrowing habit, and markedly
contrasted with the long tail and large tufted ears of
the common squirrel. Like other steppe rodents the
susliks are social and burrowing animals, hibernating in
winter, but storing in autumn a large collection of roots,
seeds, berries, &c., for winter use. Like the squirrels,
they are not averse to a certain amount of animal food,
taking small birds and their eggs when occasion offers,
and also small rodents belonging to other genera.

The marmots (Arctomys) form an interesting genus,
with representatives both on mountains and in the
steppes. We shall see later that it is not unusual to find
relationships between steppe and mountain animals,
a fact which has a double significance. In the first
place, just as the tundra forms _seppe because it is
too remote from the Equator for tee growth to occur,

(‘soufiog hop “q fq ofoyg) “yIeq oUOYSMOTTEX OY} UT MOS UeOLOMIY -g ‘OLE

‘Morn SH Jo YNoW oy} 4% , Sop , Jo youre ouperg *6 “OT

CP's" a ‘Roaung yoorBojorg 949 Aq 00YZ)

STEPPES OF ASIA AND AMERICA 63

so the upper parts of mountains form steppes because
the elevation prevents forest growth. Again, it is
believed that in those parts of the world recently sub-
jected to glaciation there was a period, after the retreat
of the ice, when the climate was unsuited to trees, lead-
ing to a wide extension of steppe conditions. This
brought in its turn a wide extension of steppe animals,
which, with the gradual improvement of the climate, and
the return of forest-haunting forms, were driven up
the mountain sides to the Alpine regions. If this be so,
then the shrill whistle of the Alpine marmot, so common
a sound in the mountainous regions of Switzerland in
summer, not only recalls the conditions of the Asiatic
steppes, but also brings back a time when much of
Europe was in the condition in which the steppes are
to-day. The Alpine marmot is now limited to the Alps,
Pyrenees, and Caucasus, the time being far past when
it also inhabited the low ground. On the other hand,
the bobac (A. bobac), the characteristic marmot of the
Asiatic steppes, extends into Europe as far as the
eastern frontier of Germany.

Marmots are rather clumsy animals with heavy
bodies, short limbs, and usually short tails. The short-
ness of the limbs must greatly limit the range of vision,
and the animals have very markedly developed that
habit of sitting up on the hind limbs to take a wider
view which is common in relatively short-legged animals.
The shortening of the fore-limbs here gives greater
purchase in burrowing, the burrows being extensive.
The food appears to be entirely vegetable, and the
animals hibernate much more profoundly than the
susliks. Bobacs occur in very large colonies, and
tunnel the ground in all directions with their burrows.
According to Kobelt, the result of these excavations is

64 STEPPE FAUNAS AND TEMPERATE

to modify the condition of the soil, so that special
plants appear in the ‘marmot gardens’, which can
thus be recognized from afar. The hemp-leaved nettle
and wild rhubarb occur especially here, and the presence
of the large leaves of these plants among the lyme
grass (Elymus) of the steppes is a very obvious feature,
while the loosening of the soil by the burrows makes
progress for horsemen very difficult over the ‘gardens ’.

The interesting point is that rodents of similar habits
seem to inhabit steppe regions in general. Thus the
prairie marmot (Cynomys ludovicianus), a related. form
inhabiting the prairies of North America (Fig. 9), has
similar habits, while the viscacha of the pampas of
Argentina, though not very. nearly related, shows also
a general resemblance in habits, and also modifies the
herbage in the regions of its occurrence.

A different, though no less perfect, adaptation to
steppe life is shown by the jerboas or jumping mice.

In speaking of the marmots we noted that for rapid
and efficient burrowing, short fore-legs are necessary,
but, on the other hand, if the hind-legs are also short,
the animal’s range of vision is small, and its speed
cannot be great. On the other hand, if the legs are
elongated to give speed, then feeding on the ground
and drinking become difficult, as is seen in the giraffe
(Fig. 36). The jerboas show a curious form of compro-
mise to these varying needs, generally parallel to that
shown by the kangaroo, which is, however, not a
burrowing animal. In the jerboas the fore-limbs are
short, thus permitting the animal to burrow, but the
hind-legs are enormously long (Figs. 37 and 38), and the
tail also long and capable of helping in the support of
the body. The animal therefore progresses by leaps, the
fore-legs not touching the ground at all. In feeding,

STEPPES OF ASIA AND AMERICA 65

a crouching position is adopted, the food being carried
to the mouth by the fore-limbs. As in rodents in
general, the food consists largely of roots, seeds, berries,
&c., and not wholly of grass, as in the kangaroo and in
most ungulates. The curious elongation of the hind-legs
gives such speed that the jerboas can escape their
enemies by flight, while on the other hand the bobacs
at once seek their burrows on an alarm, and make
underground tunnels to the feeding-grounds, so that
they are never far from an open burrow. Another very
interesting structural point in regard to the jerboas is
the reduction of the number of toes from five to three
on the hind-foot. Among the ungulates we find a
similar reduction, carried however much further, for
the horse and its allies have only one toe on each foot.
This reduction gives speed, and is an adaptation to
swift movement over a relatively hard surface.

The commonest jerboa of the steppes is Alactaga
decumana, the five-toed jerboa as it is called, because
the lateral toes are present on the hind-feet, though
they are minute and functionless. Another peculiarity
of the hind-foot, which it shares with other jerboas, is
the fusion of the bones which form the sole of the foot
in man, that is of the metatarsal bones. As a result
a very strong bone called the cannon bone is formed,
quite comparable to the cannon bone in the antelopes
and their allies, though it is formed in a different way.
This also is an adaptation to ensure swift movement
over firm ground, for it gives the necessary rigidity
during the taking of the long leaps. Though the head
and body together of this animal only measure some
seven inches, it is stated that it cannot be overtaken
by a horse, so extraordinarily rapid is its speed. Like
the other steppe rodents the jerboas are social, forming

1404 E

66 STEPPE FAUNAS AND TEMPERATE

burrows in companies. With the vegetable food they
mingle insects and the eggs and young of birds,
especially of the steppe lark.

The jerboas of the genus Dipus chiefly inhabit the
hot deserts further south, but some species, notably
D. sagitta, extend into the steppes. Though the habits
of the members of this genus are generally similar to
those of the preceding in their coloration and in their
intolerance of rain or damp, they are essentially in-
habitants of arid regions (cf. p. 133).

In the steppe region are also found the hamsters
(Cricetus), with a western extension into Europe. But
though originally doubtless steppe animals, the ham-
sters, like the brown and black rats, though to a less
extent, have attached themselves to man, and find it
more profitable to plunder his fields and gardens, than
to depend upon the precarious vegetation of the steppes.
They are remarkable for the size and elaboration of
their burrows, in which they store quantities of corn,
&c., for winter use.

Another characteristic genus of the steppe is Lagomys,
the genus which includes the picas, tailless hares or
whistling hares as they are variously called. These
animals are smaller than a rabbit, and are for the most
part mountain animals, living high up among rocks
and stones in the mountains of Central Asia. In Siberia,
however, they are found on lower ground, and even
extend into the tundra region. They form a very
important part of the food of the carnivores of the high
steppe, replacing here the lemming of the tundra. In
Mongolia, L. ogofona occurs in large colonies, and at
the approach of winter stores up large quantities of
hay near the openings of the holes. According to
Prejevalski, these haystacks may weigh 10 kilogrammes

STEPPES OF ASIA AND AMERICA 67

(22 pounds), and are sufficiently numerous for it to be
worth while for the Mongolians to bring their cattle to
feed upon the pica’s stores, and so dispense with the
trouble of storing fodder on their own account. Like
some other of the steppe animals, the picas are remark-
ably resistant to thirst, and can live in localities far
from water, where no rain nor dew occurs for months
at a time.

As to the carnivores of the steppe, the tiger, though
usually regarded as a tropical animal, extends far
to the north, and occurs in steppe regions, perhaps in
more than one variety. When nobler game fail, it does
not disdain the rodents, notably the picas. The special
cat of the steppe is, however, the manul cat (Felis
manul), a beautiful animal with a bushy tail, about the
size of the common cat, and apparently allied to the
wild cat of Europe, but, unlike it, a steppe animal. Its
main food consists of the steppe rodents, and also of
the ground-nesting birds. The steppe has also its
special fox, the corsac fox (Canis corsac), which also
feeds chiefly upon the steppe rodents. Here as else-
where the ubiquitous wolf also occurs.

In regard to the birds of the steppe it is noticeable
that they tend to acquire swiftness in running, thus
showing parallelism with the mammals. This is
carried further in the birds of the warm deserts of the
world, where we have in ostrich, rhea, emu, &c.,
examples of birds without any power of flight, but of
great fleetness of foot. The reason for this tendency is
possibly that the steppe or desert does not, like the
forest, offer safe places upon which to rest. As the
birds are exposed to danger so soon as they alight,
which is necessarily upon the ground, fleetness is a great
advantage to them.

E2

68 STEPPE FAUNAS AND TEMPERATE

As examples of steppe birds, we may name first the
bustards, of which the largest form, the great bustard
(Otis tarda), has been compared in build and habits to
the ostrich of the Saharan desert. Bustards, though
good fliers, are fleet of foot, and show a preference for
open country. They do not seem to drink, and take
both animal and vegetable feod. In addition to the
form mentioned, the little bustard (O. tetrax) and the
ruffed or Macqueen’s bustard (Hubara undulata) also
occur, the latter preferring running to flying, and
having the habit, common among steppe animals, of
squatting down when undisturbed, so as to take full
advantage of its protective coloration, but stretching
itself to its (considerable) height on an alarm in order
to increase the range of vision over the plain.

Pheasants, which are bush-haunting birds, just enter
the region, but Pallas’s sand-grouse (Syrrhaptes para-
doxus) is characteristic, both the young and the adult
showing a delicately patterned type of coloration which
is eminently fitted to conceal them in their natural
habitat. The young are precocious, being able to run
as soon as hatched. This is necessary in birds whose
eggs are laid in a mere depression scratched in the
loose soil. In the steppes the bird is said to feed chiefly
on the seeds of a member of the Chenopodiaceae called
Agriophyllum, and it assembles in large flocks in places
where this plant of the salt wastes is abundant. The
plant is also relished by horses and camels, and its
seeds are used as food by the nomads as well as by the
sand-grouse. The sand-grouse shares with many steppe
animals the peculiarity of being periodically very
abundant, when it makes incursions on lands adjacent
to its normal habitat. Flocks of it have appeared at
various times in Europe, reaching even the British

STEPPES OF ASIA AND AMERICA 69

Isles. The bird differs from the bustard in that water
is necessary to it, and it makes regular daily visits to
particular drinking-places.

Very characteristic of the Asiatic steppes are the
birds belonging to the genus Podoces, sometimes called
chough-thrushes, which are members of the crow
alliance, though their exact position is doubtful.
Though capable of flight, they only fly with reluctance,
but run swiftly, feeding upon insects and seeds, and
nesting in bushes or sometimes on the ground. They
haunt the desert regions where bushes of saxaul and
tamarisk grow, and do not seem to drink, while the more
grassy regions of the’ steppes often swarm with the
Siberian lark (Melanocorypha calandra).

In the American prairies game birds are represented
by the prairie-hen (Tympanuchus americanus), a form
related to the grouse; also by the sage-cock (Centro-
cercus urophasianus, Fig. 10), which feeds upon ‘sage-
brush’ (Artemisia tridentata) in the western states, just
as Pallas’s sand-grouse feeds upon Agriophyllum in the
Asiatic steppes ; and by the prairie-chicken (Pedioecetes
phasianellus). All these can fly rapidly and powerfully,
but only rise when hard pressed, preferring to run along
the ground, or squat among the herbage, as means of
escaping notice.

In the steppes of both the Old and New Worlds such
forms as curlews, plovers, snipe, ducks, geese, and so
on, with many of the smaller singing-birds, are abun-
dantly represented in the breeding-season, but do not
show any very notable adaptations to steppe life.

Reptiles in the steppes and adjacent deserts are not
numerous in species, but the individuals are often very
abundant. In that part of the desert of Gobi which is
called Alashan, lizards belonging to the genera Phryno-

70 STEPPE FAUNAS AND TEMPERATE

cephalus and Eremias are very abundant, and show
special adaptations to desert life, e.g. in coloration, in
swiftness, in the power of burrowing, and so forth. So
abundant are they in summer that they attract flocks

of the demoiselle crane (Grus virgo), which feeds upon

Fic. 10. Sage-cock. (Photo by the Biological Survey, U.S.A.)

them. They are also greedily consumed by birds of
prey, and even by such mammals as wolves and foxes
and domesticated dogs. In the steppes and deserts of
Turkestan there occurs a curious gecko (Teratoscincus
seincus), which has lost the disks on the digits by which
its allies climb, and has acquired curious fringes, which
enable it to run over the loose sand. Among snakes
we find the small sand-snake (Hryx jaculus), a harmless

STEPPES OF ASIA AND AMERICA 71

form, which extends into Greece, and is interesting
from the way in which it buries itself in the sand by
means of its snout ; and the poisonous T'rigonocephalus
halys, dangerous to cattle but not to man, as well as
vipers and others,

Amphibians are limited to the regions in the vicinity
of water, and offer no special features. As regards
insects we need only note that grasshoppers and locusts
are very abundant, open grassy plains being the natural
home of these insects, from which the locusts make
periodical raids upon cultivated lands.

REFERENCES. The works of Kobelt and Brehm already mentioned
give accounts of steppe animals, and reference should also be made to
any of the natural histories, e.g. the relevant volumes of the Cambridge
Natural History (London), edited by Shipley and Harmer, or The Royal
Natural History (1893-6), six vols., edited by Lydekker, where details
will be found of the different animals mentioned. A series of articles
in the American Naturalist, especially ‘Cursorial Adaptations in
Mammals’, by Lull (1904), and ‘ Fossorial Adaptations in Mammals’,
by Shimer (1903), will be found interesting in this connexion. Of
the various books published as the result of travels and giving accounts
of the life of the steppes, mention may especially be made of Prejevalski,
Wissenschaftliche Resultate der nach Centralasien unternommenen Reisen
(Leipzig, 1889). See also the books mentioned at the end of chapter iv.

CHAPTER IV
MOUNTAIN FAUNAS

Tuer animals of mountain and plateau regions show
some interesting conditions which make it worth while
to consider them separately. Taking mountains first,
we may note that elevated ranges may be divided into
three regions, each of which presents peculiar biological
features. The lowest region is usually clad in forests,
conifers predominating as we ascend. At a point which
varies with the exposure, the slope, and other factors,
even on a single mountain, tree growth ceases, and there
comes a zone with steppe characters. The width of
this band and its nature vary greatly. In e.g. the
mountain regions of Europe, it is characterized by the
growth of relatively small herbaceous plants, often with
large and gorgeously coloured flowers, but with tufted
leaves and other indications of checked vegetative
growth. Many of these plants reappear in the Arctic
tundra, but the fact that they are specially abundant
in the Alps has led to them being called Alpine plants,
whether they occur in the Alps proper, on the hills of
Scotland or Norway, in the Himalayas, or in the Cau-
casus.

In tropical regions, e.g. in Africa, this steppe area
with its Alpine plants is replaced by a band where the
plants show desert characters ; the Lobelias, Senecios,
and tree-heaths of the upper zone of Ruwenzori and
Mount Kenya form good examples. In all regions where
the mountains attain sufficient elevation, above the
steppe or desert zone comes that of eternal snow, where

MOUNTAIN FAUNAS 73

the exposed rocks are at most scantily clad in mosses
and. lichens.

Without discussing in detail the characters of moun-
tain climates, we may note that it seems probable that
the upper limit of the forest corresponds roughly to
the zone of maximum precipitation. In ascending
a mountain the rainfall increases up to a certain point,
beyond which it begins to diminish. With certain
modifications, due to inversions, &c., the temperature
diminishes steadily as we ascend. The relatively
narrow alpine zone on, e.g., the Alps is due to the fact
that the vertical distance between the zone of maxi-
mum precipitation and the snow-line is short. In
tropical regions the snow-line is pushed far upwards,
much above the zone of maximum precipitation, and
the wide interval thus produced between the forest and
the snow allows for the development of a broad band
of gradually increasing aridity, where the surface is
clad in plants of definitely drought-resisting type, the
low temperature increasing the need for xerophytic
characters by diminishing the power of absorbing
water.

Plateau regions, such as the great plateau of Tibet
and the far less extensive plateau of Spain, may be said
to correspond to the steppe zone on an ordinary moun-
tain. Plateaux of great altitude and great horizontal
extension tend to show some of the characters of a cold
desert, thus giving us the special conditions which reign
in Tibet, but the conditions tend to be less extreme
than in small and isolated areas of great elevation.
Plateaux differ from other steppes or deserts in the
peculiar nature of the relief, which demands special
agility upon the part of the animals inhabiting them,
but gives to those possessing such agility a certain

74 MOUNTAIN FAUNAS

security, due to the rugged nature of the surface.
Further, the peculiar climatic conditions, together
probably with the rapidity of mechanical erosion,
promote the growth, in certain localities at least, of

Fic. 11. A Swiss Alp. The absence of trees, the presence of a close
short turf and of low bushes (mostly berry-bearing), are very charac-
teristic of the upper regions of the Alps as well as of many other
mountain and plateau regions. The richness of the pasture is suggested
by the number of cows

particularly nourishing fodder plants. A large part
of the pastoral industry of Switzerland is due to the
pasturage obtainable on the high mountain shelves
or alps. In Tibet, though wild herbivores exist in
flocks of thousands, according to some observers, yet
there is pasturage enough to spare to feed the domes-
ticated animals upon which the community largely
depends.

MOUNTAIN FAUNAS 75

The relatively rich pasturage, and the relative
security, account for the fact that mountains and
plateaux have a considerable number of peculiar herbi-
vores, especially ungulates. The ungulates of mountains
belong to more than one group, but in the Old World
at least the two genera of sheep (Ovis) and goat (Capra),
which appeared late in geological time, include animals
which are especially adapted for mountainous districts.
Just as deer are typically forest animals, and antelopes
typically steppe or savana animals, so sheep and goats
are the typical inhabitants of elevated regions. In the
case of sheep, at least, domestication has produced so
many changes in character, that it is somewhat difficult
to realize the extraordinary agility of the wild animals,
and their other special adaptations to life in mountainous
and elevated regions. The thick coat protects them
against cold, and the fact that the young—necessarily
few in number in animals with such a habitat—are
able to follow the mother anywhere within a very
short time after birth is an important adaptive char-
acter. Like steppe animals in general, wild sheep are
social, and like many such animals they seem mostly
to appoint sentinels while feeding. Their powers of
leaping are very great, and they seem able to scale
apparently inaccessible cliffs.

While wild sheep, like domesticated forms, chiefly
graze upon the ground, the equally agile goats depend
largely upon the shoots, leaves, and twigs of bushes
and small trees, and as the trees in their native habitats
are mostly xerophytic, it is noticeable that the goats
have very catholic tastes, not disdaining the resinous,
hairy, or spiny plants-native to high steppes. From
the difference in diet the goats usually occupy ground
, which is more scarped and rocky than that favoured

76 MOUNTAIN FAUNAS

by sheep, and they have apparently even greater
powers of climbing and leaping.

As will be explained later, other types of ungulates
also show adaptations to mountain life, especially in
those regions which have no wild goats or sheep. Just
as the two latter types of animals are of relatively
recent origin, so it is probable are the adaptations to
mountain life shown by other mountain ungulates.
Indeed the fleeing to the mountains appears to be the
latest phase in the long struggle between herbivores
and carnivores. There is evidence that the herbivores
acquired swiftness as the carnivores acquired intelli-
gence and strength. As a general rule the latter seek
to capture their prey by ruse rather than by direct
pursuit, and in the open steppes and plains their task
is relatively easy. To escape them, apparently, no less
than to utilize the rich mountain pastures, some herbi-
vores sought the heights, and there the carnivores can
hardly be said to have followed them, there being few
large flesh-eaters at great elevations. Indeed the chief
flesh-eating animals of the mountain ranges of the
globe are the vultures and their allies, always con-
spicuous in such regions.

For convenience’ sake we shall take a rapid syste-
matic survey of the animal kingdom in studying moun-
tain animals, but it may be noted first that as forest,
steppe, and tundra are all represented on mountains,
it is natural that animals from all these regions should
occur there.

For example, in the Himalayas we find that those
truly arboreal animals, the Primates or monkeys, are
represented high up in the region of coniferous forests,
where, doubtless, like the ungulates, they find relative
security from the great carnivores. The characteristic

MOUNTAIN FAUNAS 77

monkey of the Himalayas is one of the langurs, a group
of slender-limbed, leaf-eating forms widely distributed
over South-eastern Asia. The particular species is
Semnopithecus schistaceus, and the animal occurs at
heights up to 12,000 feet. All who have seen it in
its native home speak of the peculiar effect of the
animal among the snow-covered pines and deodars
there, and the number of forms found on the lower
ground in India and the adjacent region proves that
the acquisition of the mountain habitat in this case
must be recent. The Himalayan form is very closely
related to the common sacred monkey of India, and
while sharing with its allies the usual adaptations to
arboreal life—to be described later—it does not seem
to show any special fitness for mountain life. Better
adapted for life at high elevations is the Tibetan langur
(S. roxellanaé), which lives in the forested regions of the
east of Tibet, especially in the vicinity of Lake Kuku
Nor, and also extends into China. It is found at
elevations of over 10,000 feet, and migrates to some
extent according to the seasons, but never descends to,
the region of cultivated land, the district plundered
so frequently by the low ground forms. The Tibetan
langur is somewhat stouter in build and shorter in the
limbs than the other forms, and has a curious tip-
tilted nose, whose use is unknown, unless it has any-
thing to do with facilitating respiration at these high
altitudes. The hair is very long, and the skin is greatly
prized by the Chinese on this account. The animal is
stated to live chiefly on ‘fruits’, but no doubt the
seeds of the mountain trees are also important. As
the other monkeys of China are species of Macacus,
there can be no doubt that the Tibetan form has found
its way into China from the south, through the moun-

78 MOUNTAIN FAUNAS

tain barrier. Its presence is thus an indication that
mountains do not necessarily form a barrier to dis-
tribution, even to animals reputed to be so sensitive to
cold as monkeys.

The macaques also include one mountain form,
Macacus tibetanus, which seems to occur in much the
same region as the Tibetan langur, and also extends
into China proper, in Sechwan and the south of Kansu.
The narrow and broken mountain chains of this region,
according to recent observers, have a damp climate,
quite different from that of Tibet proper, and into the
valleys the monsoon forest insinuates itself, taking
with it part of its characteristic fauna, notably the
two monkeys named. The macaques differ from the
langurs in their diet, for they eat insects and other
forms of animal food in addition to fruits and seeds,
As this food is much more portable tham leaves, they
have cheek-pouches in which it may be stored, these
being absent in the langurs, which have large saccu-
lated stomachs, comparable to the stomachs of some
ungulates.

The Himalayan and Tibetan langurs and the Tibetan
macaque appear to exhaust the members of the Primates
which haunt mountains. None of the New World forms
seem to be specially fitted for life at high altitudes,
and though some of the African baboons inhabit rocky
regions at some elevation, none seem to be peculiar to
high mountains or plateaux.

The insectivores have various mountain representa-
tives, some showing peculiarities of distribution. The
Alpine shrew (Sorex alpinus) inhabits the upper parts
of the lofty mountain regions of Central Europe ; the
Himalayan swimming shrew (Chimarrogale himalayica)
inhabits the lower slopes of the Himalayas, where it

MOUNTAIN FAUNAS 79

haunts streams and feeds upon aquatic insects and
their larvae, tadpoles, and small fish; the desman
(Myogale pyrenaica) of the Pyrenees inhabits the banks
of streams in that mountain range, and spends a con-
‘siderable part of its time in the water; the Tibetan
mole-shrew (Uropsilus soricipes) is a small cursorial
animal, seeming to be intermediate between moles and
shrews. The insectivores are, however, losers in the
battle of life, and there is nothing remarkable in the
fact that peculiar forms are found in mountain ranges,
when we remember the shifts the members of the group
have been put to in order to survive in a world where
the mammals as a whole have better developed brains
and greater differentiation of structure than they can
boast of.

Carnivores, aS we have seen, are not very abundant.
The forms which inhabit the neighbouring regions—
forest or steppe—may extend also into the mountain
regions, but peculiar forms are not numerous, the bear
alliance being that with most mountain representatives.

Of the cats the most characteristically mountain
form is the ounce or snow leopard (Felis uncia, Fig. 12),
a beautiful animal with very thick fur, greyish above,
with black spots, and pure white below. In Ladak it
is said to ascend as high as 18,000 feet in summer time,
and not to descend lower than 9,000 feet in winter. It
attacks wild goats and sheep, and also rodents, but its
habits are not well known. In the New World the
puma (Felis concolor, Fig. 26), though not specifically a
mountain animal, attains a considerable height in
the Rocky Mountains, and is stated to prey upon the
Bighorn sheep.

Of other cat-like carnivores extending into moun-
tainous regions, we may note that the lynx of the

80 MOUNTAIN FAUNAS

northern forests extends in a special variety, or species
(Lynx isabellina), into the plateau of Tibet, where it
shows some adaptations to rocky and barren country,
instead of to forest-covered ground. The Indian wild
dog (Canis deccanensis) shows a similar extension of
habitat from the forests of India into the wastes of
Tibet. It is a social animal, the packs hunting down
wild sheep, antelopes, &c. In the Himalayas and Tibet
also, the common fox is replaced by a variety with
thick fur and a very large brush.

We come next to the bears, of which several occur in
mountain regions. Thus the Himalayas and Tibet have
special varieties of the brown bear, which, like bears in
general, is very resistant to cold on account of the
thickness of its coat. Its very indiscriminate appetite
also makes it possible for it to exist on many different
kinds of ground. The Atlas Mountains have a closely
related form, and in the Rocky Mountains the large
and powerful grizzly bear (Ursus horribilis) occurs at
considerable elevations, though also descending to the
plain. In the wooded regions of the Himalayas occurs
the black bear (Ursus torquatus), chiefly a vegetable
feeder and an excellent climber, which in the Hima-
layas does not voluntarily quit the forest. It is not
known in Tibet, and differs from most bears in the
short and thin coat. The necessary resistance to cold
is apparently obtained by the great oiliness of the skin,
especially in autumn, when the animal is very fat—
a means of providing for the winter time of scarcity.
The Andes of South America contain another mountain
bear, in the shape of the spectacled bear (Ursus
ornatus). Africa south of the Sahara has no bear.

That forested region of Eastern Tibet which lodges
the Tibetan macaque and the Tibetan langur has still

(som M Ae adingoud oy; wot) “predoaT MOUg ‘ZT ‘DIT

)

(Photo by Capt. F. M. Bailey.

sa
a
oc
eS
s
n
~~
a
ca]
o
=)
3
a
a
ic
Pp
8
3
a
8
g
)
A

Fie. 13.

MOUNTAIN FAUNAS 81

another remarkable form in the parti-coloured bear
(Aeluropus melanoleucus). It is curiously marked in black
and white, has a very thick close fur, and enormously
powerful jaws. The latter feature is remarkable in view
of the fact that the animal seems to be purely vegetarian
in tastes. Itis said to inhabit the bamboo thickets which
in this region ascend the damp valleys up to a height
of nearly 10,000 feet. The young shoots of bamboos
form the greater part of its food, and it is stated to be
an excellent climber.

A curious little animal called the panda (Aelurus
fulgens), once believed to be related to Aeluropus, occurs
at considerable elevations in the south-eastern Himalayas
and feeds chiefly on vegetable matter, especially bamboo
shoots. The panda is an arboreal animal, somewhat
fox-like in appearance, and is now believed to be most
nearly related to the carnivorous raccoons of America,
which are forest animals, sometimes, as in the case of
the forms called coatis (Nasua), ascending to a consider-
able elevation above sea-level.

The comparative rarity of large and fierce carnivores
helps to explain, as we have seen, the abundance of
ungulates in mountain regions. Beginning with the
cattle, we may note that just as one form—the musk-
ox—is admirably adapted for life on the tundra, so
another—the yak of Tibet—is perfectly fitted for life
at a great elevation. This animal, the Bos grunniens
of zoologists, is characterized by its long hair, and by
its great agility and hardiness. It ascends to a height
of 20,000 feet above sea-level, and while impatient of
heat, is extraordinarily resistant to cold, and is capable
of thriving on the coarsest of herbage. As is well known,
the animal is used as a beast of burden in Tibet, on

account of its great endurance. Travellers have often
1404 F

82 MOUNTAIN FAUNAS

described the way in which the animals push their way
through snow, or across glaciers, or through icy glacial
torrents. They present a remarkable instance of
adaptation to very unfavourable conditions, but in this
connexion it is important to remember the great extent
of the plateau of Tibet, and its great uniformity.

The yak is the only member of the cattle group with
special adaptations to mountain life. Passing to the
sheep we find that there are eleven well-defined wild
species. Of these, eight inhabit various mountain chains
and plateaux in Asia, one, the Indian urial, extending
to relatively low ground in the Punjab and Sind ; one
form is found in the Rocky Mountains, but this, the
Bighorn, is so closely related to the wild sheep of
Kamchatka as to suggest a former land connexion
across the Bering Strait. Of the remaining two species,
one, the mouflon, occurs in mountain regions in Sardinia
and Corsica, and the other, the Barbary sheep, is found
on the southern slopes of the Atlas Mountains in Africa.
We are thus justified not only in saying that sheep are
definitely mountain animals, but also that the home of
the genus is the mountainous regions of Central Asia,
only the urial extending south of the Himalayas. The
habits of all are generally similar, and have been
already alluded to. To the preceding description we
may add a brief note on the wild sheep called argali, of
which one form (Ovis ammon) is found at moderate
elevations in Mongolia, while the other (Ovis hodgsoni)
occurs at great heights on the Tibetan plateau. Both
are large animals, reaching the size of a small donkey,
and. having short and close hair. The food consists of
grass, mingled in winter with mosses and lichens, and,
like the reindeer of the tundra, the animals in winter
seek exposed positions where the wind sweeps away

Fic. 14. The Alaskan Wild Sheep (Ovis dail), a form closely related both
to the American Bighorn and to the wild sheep of Kamchatka. (Photo by
the Biological Survey, U.S.A.)

Fic. 15. A young Takin in captivity in Tibet. The large lateral hoofs are
a special feature. (Photo by Capt. F. M. Bailey.)

Fic. 16. Half-wild Kiang, near Gyantse, Tibet. The type of country is very
characteristic. (Photo by Capt. F. M. Batley.)

MOUNTAIN FAUNAS 83

the snow. In the male the horns are large and massive,
while the female has smaller and thinner structures.
Like all their allies, the animals are very wary and
agile.

The goats are mostly distinguished from the sheep
by the presence of a beard on the chin, and by the
strong odour of the males. They also are mountain
animals, their distribution showing a remarkable
analogy to that of the sheep, except that there is no
true American goat. Some ten species of wild goat
occur in the Old World, and of these, two species, the
ibex of Abyssinia and the Arabian wild goat, which
extends into Upper Egypt, enter Africa. The remainder
occur in the mountain regions of Europe or of Central
Asia. In Europe there is a curious tendency for the
(isolated) mountain chains to have peculiar species of
goats. Thus there is a Spanish wild goat, sometimes
called an ibex; the Alpine ibex, now exterminated as
a wild animal, is peculiar to that chain, and no less
than three species of wild goats inhabit the Caucasus.
Of the Asiatic species the most attractive is the markhor
(Capra falconeri) of the Himalayas, with long spirally
twisted horns in the male, and fringes of hair on the
chest and shoulders in addition to the beard. Closely
related to the goats are a series of smaller genera, such
as Hemitragus, including the tahr of the Himalayas,
and another species found in the Nilgiri Hills of Southern
India, there being thus a curious separation between the
range of the two species. The gorals (Cemas) and
serows (Nemorhoedus) are Himalayan and Tibetan
forms, which seem to connect the goats with the
antelopes, while the takin of Eastern Tibet is a large
form with curiously shaped horns. The interest of
these genera is merely that they emphasize what has

F 2

84 MOUNTAIN FAUNAS

been said already as to the number of kinds of ungulates
found in Tibet, the Himalayas, and adjacent regions.

Though America has-no true goat, there occurs in
the Rocky Mountains a white ungulate, about the size
of a large sheep, and apparently related to the serow,
often called the Rocky Mountain goat. Its hair is very
long, and the animal seems to be very resistant to cold,
spending its time near the upper limit of the forest,
although it also descends at times to sea-level. Nearer
the antelopes than the goats is the chamois (Rupi-
capra tragus), a rather clumsily built animal with short
hooked horns, stumpy tail, and feet especially fitted
for scrambling about rocks. It is widely but discon-
tinuously distributed over the mountain regions of
Europe, but is found fossil on the low ground also.

Of the true antelopes many of the gazelles ascend to
considerable elevations, notably the Tibetan gazelle
(Gazella picticaudata), which ranges up to 18,000 feet,
and is less shy than most gazelles. Tibet has also
a peculiar antelope, the chiru (Pantholops hodgsont), an
animal in which the male has a curious swollen nose
and long horns. This form occurs at the same heights
as the gazelle, sometimes in very large herds. In the
distended nose it somewhat resembles the saiga of the
Siberian steppes.

Of the vast number of African antelopes some
frequent hill regions, notably the rhebok (Pelea cap-
reola) of the south and east, which has the habits of
a chamois. Another agile and chamois-like African
form is the klipspringer (Oreotragus saltator), found at
heights up to 8,000 or 9,000 feet in Abyssinia, and
having very small feet, which apparently give it
great sureness on the rocks. Both its Dutch and Latin
names refer to its jumping powers.

Fig. 17. The Bharal or Burhel, one of the wild sheep of Tibet (Ovis nahura).
Note the curious markings on the legs. (Photo by Capt. F. M. Bailey.)

Fie. 18. The Serow of Tibet. (Photo by Capt. F. M. Bailey.)

Fic. 19. Musk-deer. Note the large lateral hoofs, the coarse coat, the prominent
canines of the male. (From a specimen in the Royal Scottish Museum.)

MOUNTAIN FAUNAS 85

Of the deer we may note that the somewhat primitive
musk-deer (Moschus moschiferus) occurs at consider-
able elevations in the Himalayas, being usually found
in the birch forest above the zone of pines. From the
Himalayas the musk-deer extends northwards into
Central Asia and Siberia. It has unusually large lateral
hoofs, which apparently assist it in obtaining foothold
on hard snow or rocky ground, and feeds largely on
twigs and buds. The coat is thick, the hair being long
and coarse, and giving much protection against cold.
The kiang (p. 60) occurs in Tibet as well as in the
Mongolian steppe.

As already explained, South America is: singularly
poor in ungulates, but in the Andes two members of
the camel family occur, the guanaco and the vicuna.
These animals have very soft feet, and though they
ascend to the highest ridges, they avoid equally rocky
regions and snow and ice. At the wet season they go
high up the mountain sides, but when the vegetation
there dries up at the approach of the hot season they
descend to valleys watered by springs or perennial
streams. The long coat affords protection against cold,
and the animals seem never to descend below 6,000
or 7,000 feet in the hotter parts of their range.
That this is partly due to intolerance of heat seems
indicated by the fact that the guanacos come down to
sea-level in Patagonia. The vicuna has a more limited
extension, being confined to the district between
Southern Ecuador and Central Bolivia. One of the
American tapirs also (Tapirus roulint), ascends to an
elevation of 7,000 to 8,000 feet above sea-level in
Ecuador. and Colombia, without, apparently, showing
any special adaptations to mountain life. In the
Andes of Chile and Ecuador also occur the tiny deer of
the genus Pudua, which are no bigger than hares.

86 MOUNTAIN FAUNAS

The very aberrant and somewhat primitive ungulates
called hyraces (Procavia) show some interesting features
as regards adaptation to mountain life. All are small
animals, very like rodents in appearance and habits,
and resembling most of those animals also in their
helplessness and want of swiftness. When small herbi-
vores cannot defend themselves actively, cannot escape
by flight, and cannot burrow, only two possible means
of escaping their enemies remain. These are to haunt
rocky country where the fallen stones and blocks offer
natural shelter, or to seek the trees. Both methods are
adopted by the hyraces. The common hyrax of Syria
gets its name of rock coney from its habit of frequent-
ing rocks, where it pops in and out of the crevices as
a rabbit pops in and out of its burrow. As rocky
country is more frequent in mountainous regions than
elsewhere, we find that the ground hyraces usually
occur in elevated districts, though frequenting suitable
ground elsewhere in addition. They have four well-
developed toes on the fore foot, and three on the hind,
and the size of the feet, with the spreading toes, enables
them ‘to scramble about rocks very easily. The tree
hyraces, without any notable difference in structure,
show a marked difference in habit in that they live
entirely among trees, in which they feed, breed, and
sleep. Being relatively hardy they occur at great
elevations, e. g. at from 7,000 to 11,000 feet in the forests
of Kilimanjaro. In both the ground and tree forms,
those attaining a great elevation seem to differ chiefly
in their thicker coats. All the hyraces are confined to
Africa and the adjacent regions, e.g. Syria and Arabia.

Among the true rodents we find, as already indicated,
many mountain species. In the consideration of steppe
faunas something was said of marmots (Arctomys),

(umasnyy ysnoog yohoy oy2 ut uowroeds 0 wolg) ‘oIstioyoereyo ore ‘(gE “21,7 008)
AUopol v Jo asoyy ejqueser ATjeroyredns Yor ‘q999} AOSTOUL SNoTIno ayy, *(zv/hzZ) ouog Yooy uerwsdyg oy],

nee ae

06 “SLL

MOUNTAIN FAUNAS 87

and as these are typical denizens of cold steppes, it is
natural that they should occur also in the steppe region
of lofty mountain chains. In Europe the Alpine marmot
is found in the Alps, the Pyrenees, and the Caucasus,
though it once occurred on the low ground also. In
Central Asia many kinds of marmots occur, though
none extend to the south of the Himalayas. The
Himalayan marmot, allied to the bobac of the steppes,
is limited to the region above the tree limit, that is, to
the barren wastes which most nearly resemble the home
of the bobac, the usual elevation being 12,000 to 13,000
feet. In North America we have similarly a Rocky
Mountain marmot, resembling in habits the Alpine form.

The voles (Arvicola) are well represented in mountain
regions, the genus showing a remarkable tendency to
run into varieties, or species, in particular localities.
Thus in the Alps and Pyrenees there occurs the very
curious Alpine or snow vole (A. nivalis), which is some-
times pure white, but is structurally very near the
continental field vole and the bank vole. It differs
notably from these forms in its habits, for it lives high
up the mountains, even above the snow-line. It has
been found on the Finsteraarhorn at a height of 12,000
feet, but on account of its habits is rarely seen. It
appears to live actually beneath the snow, making
runs beneath it after the fashion of the Arctic lemming,
in its search for vegetation. Other mountain species
occur in the Himalayas and Tibet.

South America, with its paucity of ungulates, has
a considerable number of peculiar mountain rodents.
Thus the common chinchilla (Chinchilla lanigera), with
its soft thick fur, inhabits the higher parts of the Andes,
living in rocky ground, and displaying marvellous
agility. A larger form (Lagidiwm cuviert) inhabits the

88 MOUNTAIN FAUNAS

same regions, and has similar habits. The pacas of the
Brazilian woods are represented on the mountains by
a smaller type (Coelogenys taczanowskii). Some of the
cavies also ascend to mountain regions, notably the
Bolivian cavy (Cavia bolivensis), which lives in large
societies in the upper part of the Andes, tunnelling the
ground in all directions with its burrows.

As to the hare-like rodents, we have already noted
that the picas or tailless hares occur at great elevations
in the Himalayas in addition to frequenting the Siberian
steppes, thus illustrating the same conditions as the
marmots. The mountain hare of Europe is closely
related to the Arctic hare already discussed (p. 22).
Tibet has several species of hares, the most interesting
being Lepus hypsibius, which does not descend below
14,000 or 15,000 feet.

Among birds we find that many kinds of birds of
prey haunt mountains, though they are not necessarily
confined to them. Of wide distribution is the laem-
mergeier (Gypaétus barbatus), found in the mountain
ranges of Europe, North Africa, and Central Asia,
which apparently feeds largely on carrion and animals
not killed by itself. The even more widely distributed
Golden Eagle (Aquila chrysaétus), on the other hand,
kills its own prey, which consists of rodents, the young
of ungulates and game birds. Other forms which may
be mentioned are Bonelli’s hawk-eagle (Nisaétus fascia-
tus), found in the Alps as well as in Asia and North
Africa, and the magnificent condor (Sarcorhampus
gryphus) of the Andes of South America. In tropical
countries some of the parrots range up mountain sides,
and in this connexion it is curious to note that the
New Zealand kea (Nestor notabilis), which inhabits
barren mountain sides, has become virtually a bird of

MOUNTAIN FAUNAS 89

prey since the introduction of sheep into that area. At
first confining itself to dead sheep, the bird is stated now
to kill living animals with blows of its powerful beak.

As is to be expected from the conditions, game
birds are frequent on mountains. Thus in the Alps we
find the ptarmigan, and it is also found on the other
mountain ranges of Central Europe, on most of which
the black grouse or capercaillie is found. In the
mountain ranges of South Europe generally the Greek
partridge (Caccabis saxatilis) lives. To the pheasant
family belong the snow-partridge (Lerwa nivicola) of
the Upper Himalayas, and the very large snow-cock
(T'etraogallus himalayensis) of the Himalayas and Tibet.
Other examples might be given, but these sufficiently
indicate the two essential facts—that some of the game
birds of the tundra and pine forests extend to the
mountain ranges of the interior, and that the great
elevated area in Central Asia has its peculiar game birds,
as it has its peculiar ungulates and its peculiar rodents.

In regard to other types of birds, a few words may
suffice. In the Alps the commonest member of the
crow family is the Alpine chough (Graculus alpinus),
smaller than the common form, and having a yellow
instead of a red beak. In the pine forest of mountains,
especially in the Tyrol, occurs the citril finch (Chryso-
mitris citrinella), while high up on the Alps is found
the beautiful black and white snow-finch (Monti-
fringilla nivalis), allied forms of which also occur on
the mountains of Central Asia. Both in the Alps and
in Central Asia is found the curious wall-creeper (Ticho-
droma muraria), which lives on insects and spiders,
found by probing in rocky crevices. In shape and
flight this bird has a remarkable resemblance to a large
butterfly, as it hovers over the rocks, using its long

90 MOUNTAIN FAUNAS

sharp bill to pick out its prey just as a butterfly uses
its proboscis to suck up nectar. In the Alps another
frequent bird is the rock-thrush (Monticola saxatilis),
a brightly coloured bird, extending up into the barren
regions of the mountain sides. In the Alps and Pyre-
nees also the Alpine accentor (Accentor collaris) is
common. With the exception.of the hedge-sparrow,
most of the accentors are mountain-haunting birds,
having strong feet and legs, which fit them for pro-
gression over rough ground. Of the summer visitors to
the Alps, we may note the Alpine swift (Micropus
melba), which is larger than the common form.
_ Reptiles do not appear to show any special adapta-
tions to mountain life. It is remarkable that the
common viviparous lizard (Lacerta vivipara) ascends
to a height of 10,000 feet in the Alps, but then this
hardy animal also extends into Siberia. The common
viper (Pelias berus) also occurs in the Alps, though not at
such a great elevation, and it has a similarly wide range.
As regards amphibians, there is some evidence that
the mountain habitat tends to produce darkening of
the colour. Thus the common frog (Rana temporaria)
occurs in the higher parts of the Alps in a dark-coloured
variety, and the same thing is said to be true of the
common toad (Bufo vulgaris). Similarly the Alpine
salamander (Salamandra atra) is black in colour, and
unspotted, thus differing markedly in colour from its
allies. This form shows a peculiar adaptation to life
at considerable elevations, in that it only produces two
young at a time, and these pass through the gilled stage
within the body of the mother, so that at birth they
are lung-breathers like the parents, and not gill-bearing
tadpoles. As the animal ascends to an elevation of
9,000 feet, it is obvious that ice-free water for the

MOUNTAIN FAUNAS 91

relatively long period of larval development would not
be easy to find—hence the adaptation described. In
mountain streams in China and Japan occurs Siebold’s
giant salamander (Cryptobranchus maximus), reaching
a length of over five feet.

Relatively few species of fish occur in mountain
streams, doubtless on account of the coldness of the
water.

In regard to the lower forms of life we need only note
that with the great development of flowering plants
during the short period of summer in most mountain
regions, we have a great development of insects,
especially butterflies. There is a marked general
resemblance between the insects of mountains and of
polar regions, due doubtless partly to the similar
physical conditions which prevail in the two regions.

REFERENCES. Kobelt’s book has a good chapter on mountain animals,
and Tschudi’s Thierleben der Alpenwelt (ninth edition, Leipzig, 1872)
should also be consulted. An account of Alpine animals will also ‘be
found in Anderegg’s Schweizerische Alpwirtschaft (Berne, 1899). There
are a number of works on travel in Tibet and the neighbouring regions
which give accounts of various of the characteristic animals. Among
these may be mentioned: Demidoff’s After Wild Sheep in the Altai
(London, 1900); Adventure, Sport, and Travel on the Tibetan Steppes,
by W. N. Eergusson (London, 1911); P. T. Etherton’s Across the Roof
of the World (London, 1911).

CHAPTER V
THE FAUNA OF THE TROPICAL FOREST

Just as the tropical forest as contrasted with the
temperate is remarkable for its wealth of species of
trees and for the luxuriance of its undergrowth, so its
fauna is remarkable for the number of special types.
Adaptations to life in the tropical forest occur in almost
all groups of animals, and some orders, like the Primates
among mammals, are, roughly speaking, fitted for this
habitat alone, for relatively few of their members occur
outside this region.

We may begin by a few words upon the special con-
ditions which reign here. To begin with, food is abun-
dant all the year round, for there is no seasonal check
to vegetation such as occurs in temperate climates.
The absence of seasons, in the sense in which they
occur in higher latitudes, makes possible the occur-
rence of many fruit-eating animals, for fruits occur at
all seasons. Thus the anthropoid apes are fruit-eaters ;
the fruit-bats have the same diet, as the name indicates ;
parrots eat seeds and fruits, and so on. For such
forms life in colder climates is impossible.

The constantly high temperature, at least during the
day, makes a warm coat unnecessary, so that animals
producing valuable furs are relatively rare. The same
condition makes reptilian life abundant, for the heat
of the sun is available to hatch the eggs. The high
temperature also results in abundance of insects, which
again feed many other animals.

As the tropical forest is characterized by a very

FAUNA OF THE TROPICAL FOREST 93

luxuriant undergrowth, and by great density, we find
not a few arboreal forms which never voluntarily put
foot to the ground. This is true of the South American
sloths, whose feet are quite unfitted for terrestrial pro-
gression, also of some of the monkeys, of forms like the
flying lemur, and so forth. Where the forest is less
dense than usual, such purely arboreal forms must at
times find it difficult to pass from one tree to another,
and it is notable that very many different kinds of
arboreal animals have some form of parachute, giving
them partial powers of flight, or at least of taking long
leaps. Thus we have flying frogs (Rhacophorus) ; a fly-
ing lizard (Draco volans) ; flying phalangers ; African
flying squirrels, differing in several respects from the
forms found in temperate climates; the so-called
flying lemur, &c. The true lemurs do not possess
parachutes, but some forms (e.g. Galago) have the
ankle greatly elongated, which gives them a very frog-
like appearance, and enables them to leap from one
tree to another.

As regards the structure of the feet, two conditions
occur in the more highly specialized forms. Some-
times, as in the sloths of South America, the fingers
and toes are converted into hooks, incapable of being
separated from one another, and having as their sole
function the suspending of the body from the branches
of the trees. This condition is suggested also in some
of the thumbless monkeys, but in general the monkeys
and their allies, together with some other animals,
show a .character which gives as great security of
grip, while conducing to greater freedom of movement
and greater agility. This is the modification of one
digit on both hands and feet, which is so inserted that
it can be opposed to each of the other digits, producing

94 THE FAUNA OF

the condition described as the opposable thumb and
great toe. At the same time the extremities of the
digits tend to be flattened, giving more gripping power,
and with the flattening of the finger-tips a nail tends
to replace a claw. This condition is only gradually
acquired in the Primates, the lower forms having
pointed fingers, with nails instead of claws. As already
suggested also, the hand in certain Primates tends to
become hook-like, the thumb being absent or minute.
Only in man among the Primates is the great toe not
opposable.

But this condition of one (or more) digit being
opposable to the other is not confined to the Primates.
In the opossums and phalangers, primitive mammals
whose young are born in a very undeveloped state, the
great toe is also opposable. Again, in parrots among
birds, and the chameleon among lizards, a gripping
hand and foot is produced by certain digits being
opposable to the others.

Still another adaptation to arboreal life is seen in
the prehensile tail. This also occurs in widely separ-
ated groups. We find it in the chameleon, in the New
World monkeys, in opossums, in some of the South
American ant-eaters, in the South American tree-por-
cupines, in the kinkajou, and so forth. Like the hook-
shaped hand it tends rather to occur in relatively sluggish
forms than in those whose agility is sufficient to enable
them to recover easily from a false movement. Thus
Brehm notes that the ‘five-handed’ New World
monkeys are not nearly such good climbers as the
four-handed monkeys of the Old World, which never
have prehensile tails.

Some minor adaptations to the arboreal life may
next be touched upon. As great freedom of movement

THE TROPICAL FOREST 95

of the limbs is necessary, we note that these are not
tied down to the body by skin, as in most quadrupeds.
A comparison of such animals as dog or cat with
monkey or man will make this point clear. Again, as
the arms are of very great importance in arboreal
animals, we find that they are proportionately very
long. In the sloth, in the monkeys and lemurs, this is
very noticeable. With the loss of other arboreal charac-
ters man has also lost his long arms, these being pro-
portionately very much shorter than in the apes.

Another common feature of arboreal mammals is
the presence of collar-bones, which tend to be lost in
quadrupeds. This gives such mammals a charac-
teristically broad chest, easily seen on comparing the
position of the fore-limbs in, e.g. a sheep, and in a
monkey. It is the presence of the clavicle which per-
mits of the upward movement of the arm, so necessary
in climbing, a movement impossible to pure quadru-
peds like antelopes and deer. Again, in quadrupeds,
and. notably in the swift ungulates, the two bones of
the forearm and the corresponding bones of the fore-
leg tend to fuse, in order to give a rigid support to the
body ; but in most arboreal animals they are separate, -
giving much greater freedom of movement to wrist
and ankle, but preventing the animals from acquiring.
any great speed on the ground, even when their fingers
and toes make it possible for them to attempt to run.

Finally, arboreal forms have often only one or two
young at a birth, and the mammae are frequently
pectoral in position, for only thus can the young be
carried without loss of balance by the mother.

To this account of the chief peculiarities of arboreal
animals we may add a few words upon the tropical
forest considered as a source of food. In the tropical

96 THE FAUNA OF

forest there is, as already indicated, no seasonal check
to vegetation, and therefore no marked variation in
the amount of vegetable food available. But there is
much evidence to show that the total amount is less
than might be expected. The tropical forests of New
Guinea yield almost no food for man, so that recent
expeditions have had to import every particle con-
sumed, and have at times run almost the same risks
of starvation as explorers in the tundra, even when
armed with modern weapons. Elsewhere there are few
human groups indeed who manage to subsist solely on
the products of the tropical forest, and those who do,
e.g. the Congo pygmies and the Philippine negritos,
seem to be few in number, and yet to suffer from chronic
starvation. They do not reach a high state of civiliza-
tion. The fact that all the more intelligent of the
forest animals soon learn to rob man’s plantations, or
to prey upon his flocks and herds, suggests that the
wild mammals are no better off than forest-dwelling
man. ‘The reasons why the conditions should be
relatively unfavourable in the tropical forests are
perhaps a little obscure, but we note among them that
the physical conditions most favourable to plant life
are not those best suited to the mammalia at least.
Further, the enormous number of plant species in the
tropical forest diminishes the chance that a particular
useful species will occur in numbers. The fact that
trees of commercial value only occur in isolated speci-
mens is a great barrier to the exploitation of the tropical
forest by man, and it has doubtless also its effect on
the wild animals. If there is one Cembra pine loaded
with ripe seeds in the taiga, there will almost certainly
be many in the vicinity; if the squirrels find one
hickory tree in the American woods whose nuts are

THE TROPICAL FOREST 97

ripe, there will also be many others. On the other hand,
just as man has to hunt for individual camphor trees
in the Formosan forests, or for individual rubber trees
in the Brazilian forest, so must the fruit-eating wild
mammals hunt through the forest for the scattered
trees whose fruits supply their hunger. This is perhaps
part of the reason why the tropical forest seems to the
explorer to be deprived of animal life save insects and
reptiles, part of the reason why the forest mammals
mostly occur in relatively small numbers.

If we consider in systematic order the groups of
animals showing adaptation to life in the tropical forest,
we naturally begin with the Primates, whose members
display practically all those characters which we have
described. Their long arms and short legs, their oppos-
able thumbs and great toes, their strong clavicles, in
some instances their prehensile tails, their pectoral
mammae, and the reduction in the number of young
at a birth—all these fit them for life among the trees.
On the other hand, their generalized teeth, the number
of fingers and toes present, the condition of the bones
of the limbs, and other characters, show their descent
from relatively primitive ancestors, specialization show-
ing itself chiefly in brain development, and in the fact
that the hands are progressively suited for many uses,
in addition to their prime function as aids in climbing.

One point of structure is interesting. Most herbi-
vorous, or partially herbivorous, animals are specially
exposed to danger when feeding, for large quantities
of vegetable food must be taken. Many of the ungulates
have an arrangement whereby the hastily swallowed
food can be stored within the alimentary canal, careful
mastication taking place later. In most of the monkeys,
on the other hand, food can be temporarily stored in

1404 G

98 -THE FAUNA OF

cheek-pouches, to be chewed and swallowed later. This
arrangement is far from universal—it is absent in all
the New World monkeys and in some Old World
forms, but it is nevertheless common.

As, with the exception of baboons, and of the Tibetan
and Himalayan forms already mentioned, all Primates
are denizens of the tropical forest, it is impossible to
name all the members of the order here—a few ex-
amples only can be picked out.

Apart from man, the highest living forms are the
anthropoid apes, of which there are four living kinds—
the gorilla and chimpanzee from West Africa, the orang
in Sumatra and Borneo, the gibbons from South-East,
Asia. Like man himself, and like most Primates except
the lemurs, the anthropoid apes are diurnal in habit,
not nocturnal, like many mammals. The gorilla and
chimpanzee resemble man in the region which they
inhabit in having dark skins ‘and hair ; the body is of
course much more hairy than in man. The gorilla,
which has a very limited distribution, is a dweller in
dense forest regions, and is an extraordinarily powerful
animal. It is purely vegetarian and fruit-eating, shar-
ing a taste for some fruits with the negroes of the
region, and also consuming some which the negroes do
not care for. The fact that it robs human plantations
shows that food is sometimes scarce, even though the
animals appear to be very few in number. Wild plan-
tains and ‘ palm cabbage’ are freely eaten.

The chimpanzee, which extends into Central Equa-
torial Africa apparently, at least at times, eats some
animal food in addition to fruit.

Very different in appearance from the gorilla and
chimpanzee is the orang, with its very long arms and
its covering of long reddish hair. The way in which

Fic. 21. The Margin of the Equatorial Rain-Forest in North Borneo. Where-
ever this type of forest occurs it shows similar density (see Fig. 22), a fact
which explains the existence of so many purely arboreal animals (cf. Figs. 23
and 32) which spend their whole lives among the trees. (Photo by H.M. Lomas.)

Belgian Government.)

a
a
3

&

3

a
Ba
°
isl

H
°
a
i}

o
2

P=]

a

Fig. 22.

THE TROPICAL FOREST 99

the knee turns outwards makes it even less fitted for
the upright position than its allies, but the resultant
inward position of the sole of the foot makes it even
more skilful in climbing. The animals are indeed
excellent climbers, but are somewhat deliberate in their
movements. They do not appear to come voluntarily
to the ground, and are fitted for dense primaeval forests.
The gibbons are very much more active in their move-
ments ; so active, indeed, as to be able to catch birds on
the wing. They mingle with their vegetable food a
considerable amount of animal matter, especially eggs
and nestling birds. With the other egg-eating animals
of the tropical forests they have probably been one
factor in determining the migration of so many birds
every year to the northern forest, where they can rear
their young in comparative security.

The gibbons have even longer arms than the orang,
but differ in that they can plant the foot flat upon the
ground, and walk, though not rapidly, in the upright
position.

Of other forms we may note that the langurs, already
mentioned (p. 77), are long-tailed, leaf-eating monkeys,
widely distributed over South-East Asia. Though
the arms are relatively short, the animals are exceed-
ingly agile, making their way through the trees with
great rapidity. In West Africa the langurs are re-
placed by thumbless monkeys belonging to the genus
Colobus, which are also leaf-eaters, and are remarkable
for their long silky hair.

West Africa, the region, it will be recollected, where
the tropical forest is best developed, is also inhabited
by the monkeys of the genus Cercopithecus, to which
the monkey usually carried about by organ-grinders
belongs. All have cheek-pouches, and the diet is very

G2

~

100 THE FAUNA OF

varied. In Asia the macaques are similar forms, with
very large cheek-pouches and a most indiscriminate
appetite, animal food entering largely into the diet.
The baboons of Africa are much more widely dis-
tributed through that continent than most genera of
monkeys, but this is associated with the fact that they
are not forest animals in the true sense, being chiefly
found on rocky ground, and being practically quad-
rupeds. The shortness of the arms is a drawback in
climbing, at which they do not appear to be skilful.

Very different from all the Old World monkeys are
those which haunt the equatorial forests of Brazil, and
constitute the Platyrrhine or American forms. All these
animals are small, none has an opposable thumb, the
tail is usually prehensile, and the partition between the
nostrils is broad. Examples are the various kinds of
capuchins, which have a mixed diet, and share with
some Old World forms the habit of plundering culti-
vated land. In the absence of cheek-pouches they carry
away part of their spoil in their hands or under their
arms, the arms not being indispensable in climbing,
which is carried on largely by the tail. The spider mon-
keys (Ateles) are comparable to the thumbless monkeys
of Africa in the total absence of the thumb. As their
name indicates, they have long slender limbs, and are
very skilful climbers, the long and markedly prehensile
tail playing a great part in the process. The leaf-eating
howling monkeys (Mycetes, Fig. 23) may also be named,
but a considerable number of other genera occur, some
without prehensile tails. Though characteristic of the
equatorial forests, these New World monkeys extend
northwards into Southern Mexico, and southwards to
lat. 30° 8.

Of much more limited distribution are the little

Fie. 23. The Red Howling Monkey, an American form with a prehensile tail.
(From a specimen in the Royal Scottish Museum.);

(‘unasnyy ystnoog yohlowy 242 ur uauroads 0 wotg) “UMOYS TOM St
esoU pvolg eT, ‘opisueyord-uow st [req oyy Yor A UT WO; & Jo ofdurexo ue (xyjoshuyD)- AoyUop WeoWoULy UY “PZ “OL,

THE TROPICAL FOREST 101

marmosets, which occur only in the tropical forests of
South and Central America. These animals are very
squirrel-like in appearance, and show much fewer
adaptations to the arboreal life than the higher Pri-
mates, in spite of the fact that they spend their lives
among the trees. Thus the fingers and toes, except the
great toe, have pointed claws instead of nails, the hind
legs are larger and better developed than the front
ones, and the great toe is very small. The tail is non-
prehensile. In the absence of the special adaptations
shown by their allies, the animals climb like squirrels
rather than like monkeys—that is, they do not grip
the branches and swing freely from one to another,
but stick in their claws and climb along the branch on
all fours. The diet consists of fruits and insects. L
The lowest:-members of the Primates are the lemurs,
which show some marked peculiarities of distribution,
as well as of structure. They are entirely absent from
the New World, and in the Old World have apparently
been pushed far to the south by the development of
better organized forms, or by the attacks of carnivores,
for all lemurs are very helpless animals. Existing
lemurs occur in the tropical forests of Africa, in the
forests of the south-eastern region of Asia, an area
which contains many strange and primitive animals,
and finally in the island of Madagascar. In Africa and
Asia they are few in number, apparently in individuals
as well as in species, and are always small in size. In
Madagascar there are more than thirty species, and the
individuals are abundant in every wood. Here also
the animals reach a relatively large size, the largest
being about two feet in length, and, as the number of
species suggests, they show adaptations to varying con-
ditions of life. The number of species and differentia-

102 THE FAUNA OF

tion of structure seem to be associated with the fact
that the lemurs here are protected by isolation, living
as they do on an island where there are no true monkeys
nor apes, and where the carnivores of the cat family
are absent, members of the less differentiated civet
alliance taking their place.

Lemurs are less intelligent than monkeys, and much
less highly differentiated, but they are no less definitely
adapted to arboreal life. It is therefore the more
interesting to find that, just as baboons are monkeys
which have abandoned the arboreal life, so in Mada-
gascar we find the ring-tailed lemurs, foxy-looking
animals, which live among rocks in regions where trees
are virtually absent. We have already spoken of the
fact that the galagos of West Africa have an elongated
ankle which, though the mechanism is a little different,
gives them the power of leaping like a frog. The same
peculiarity occurs in the mouse-lemurs (Chirogale) of
Madagascar, and is even better developed in the little
tarsier (Tarsius) of the Malay region, which progresses
by leaping from one branch to another, or from one
end of a branch to the other.

We have not hitherto spoken of bats here, because
those forms which occur in temperate regions show no
special adaptation to any particular type of habitat.
It is otherwise with the large fruit-bats, which are
practically limited to the tropical regions of the Old
World, where they are chiefly found in forests. They
are entirely absent from the New World, but in the
east extend southwards to the island continent of
Australia and to Tasmania, though not to New Zealand.
The fruit-bats are larger than the insect-eating forms,
the common fruit or fox-bat of India measuring four
feet from tip to tip of the wing. Their mastery of the

Fic. 25. Male and female of the Black Lemur, Madagascar. The
male only is black, the female being brown with a white ruff. Note
the opposable thumb and great toe. Though the general shape is
monkey-like, the foxy faces should be contrasted with those of
monkeys (Figs. 23 and 24). (From a specimen in the Royal Scottish
Museum.)

THE TROPICAL FOREST 103

air is less complete than in the insect-eating forms, the
arrangement of the tail apparently preventing them
from giving those sharp turns so necessary in an animal
which feeds on insects caught on the wing. Further,
as the fruit-bats feed in trees they are more arboreal in
habitat, being able to scramble about the trees, attach-
ing themselves by the hind feet, and by the strongly
hooked thumb of the hand. Similarly the index finger
is not reduced to the rudimentary condition seen in
other bats, where it is useless, and is here usually fur-
nished with a claw. Fruit-bats spear their food with
the claw of the thumb, and have teeth so modified as
to allow them to crush pulpy fruits. From the diet
the animals must necessarily have been once confined
to regions where wild fruits were abundant, but, like
some members of the order Primates, they have taken
advantage of man’s fondness for fruit to extend their
range to regions where his plantations can be robbed.
In Australia, where wild fruits are somewhat rare, the v
fruit-bats seem to devour the flowers of the eucalyptus,
but the way in which they raid the orchards of the
fruit-growers shows that fruit is greatly preferred when
it can be obtained.

In the forests of tropical America there occurs an
interesting family of bats, the Phyllostomatidae, or
vampire bats, some of whose members present a curious
analogy to the fruit-eating bats of the Old World. The
vampire bats belong to the insect-eating section (Micro-
chiroptera) of bats, but nevertheless some of them eat
fruit only. Others eat fruit and insects, while others
again are purely blood-suckers.

We have already pointed out that insectivores are
peculiarly helpless animals, without the intelligence
which enables many of the Primates to escape from

104 THE FAUNA OF

the consequences of their own incapacity for defence.
The tropical forests offer great possibilities in the way
of shelter and protection, and it is therefore not remark-
able that various forms should show special adapta-
tions to life here. The most striking of these adaptations
are those shown by a very aberrant animal, the so-
called flying lemur (Galeopithecus), of which two
species occur, one in the Malay Peninsula, Borneo,
Sumatra, and adjacent regions, and the other in the
Philippine Islands. The flying lemur is about the size
of a cat, and resembles the flying squirrels of temperate
forests in having a parachute of skin extending from
the fore to the hind limbs, which enables it to take
flying leaps from tree to tree. The membrane is, how-
ever, better developed, in that it involves the tail as
well as the limbs. The fingers and toes are short and
strongly clawed, and they, together with the slightly
prehensile tail, enable the animal to swarm up trees,
for it cannot fly in the sense of ascending against
gravity. It feeds upon leaves, and has strongly cusped
front teeth, presumably for the purpose of nibbling the
leaves. The animal altogether offers a curious example
of adaptation to forest life.

In the same region as the flying lemur, but extend-
ing also into the mainland of India and Burma, we find
the little tree-shrews (Tupaia), which so far as adapta-
tions go may be said to be comparable to ordinary
squirrels, as Galeopithecus is to flying squirrels. The
tree-shrews are very like squirrels in appearance,
though smaller than most. The feet are naked beneath,
with sharp claws to allow the animals to cling to the
rough bark, the tail is long and bushy, and the animals
use their forefeet like squirrels in feeding, sitting up
on their hind-legs and holding the insects or fruits in

THE TROPICAL FOREST 105

the forepaws. Structurally, of course, the tree-shrews
are very different from the squirrels, so that the resem-
blances are due to adaptation. Their much more
limited distribution shows that they have been much
less successful than squirrels.

The powerful and intelligent carnivores have many
representatives in the tropical forests, though their
dominance makes it less necessary for them to acquire
purely arboreal habits than for more helpless animals.
They haunt the forest for food rather than for protec-
tion. Of the large forms the lion is not a forest animal,
though occasionally found in forest regions. The tiger,
in India especially, is much more frequently found in
wooded regions than elsewhere, but this may be partly
because it is very impatient of great heat. It is appar-
ently a bad climber. On the other hand, the widely
distributed leopard climbs well, being capable of run-
ning up a straight-stemmed tree. This is even more
true of the South American jaguar, which in swampy
regions may be almost purely arboreal in habitat, and
preys largely upon monkeys. It shows, however, like
the other large cats, considerable adaptability, being
found in savanas as well as in forests.

The puma (Fig. 26) does not appear to haunt dense
forests, being much more a denizen of open plains or dis-
tricts with thick reeds and grass. The clouded leopard
or tiger (Fig. 45) is an arboreal form found in South- 3
East Asia. The smaller cats, with the exception of
Madagascar and Australia, are universally distributed,
and the tropical forests of all regions have their full
share of species.

On the other hand, the less differentiated forms known
as civets (family Viverridae) are limited to the warmer
parts of the Old World. The habits of the civets are

106 THE FAUNA OF

varied, but among the arboreal forms we may note the
palm-civets or tree-cats (Paradoxurus) of India and
West Africa, which have naked feet, a common feature
in arboreal animals, and the bear-cat (Arctictis) of
South-East Asia, with a prehensile tail.

Though not a few of the dog alliance extend into
wooded country, none can be said to have any special
adaptation to forest life.

On the other hand, the bears and their allies, despite
their total absence from Africa south of the Sahara,
and their general preference for temperate and upland
regions, are well represented in the tropics. Thus the
forests of South-East Asia contain the small black
Malayan bear (Ursus malayanus), which is essentially
arboreal and frugivorous in habit. In the forests of
India the sloth-bear (Melursus ursinus) is common, and
lives in forest regions, though it is a clumsy climber.
Of the raccoons, the coatis (Nasua) are found in the
forests of South and Central America, and the kinkajou
(Cercoleptes caudivolvulus) of the tropical forests of
America, a cat-like animal, has as special arboreal
characters a prehensile tail, naked feet, and strong
claws. It feeds upon small mammals, birds and their ~
eggs, and fruits. The widely distributed weasel family
has representatives in the tropical forests, but the animals
do not predominate here as they do in the north, and
the tropical forms show no special features of interest.

As a general rule ungulates are not abundant in the
tropical forests, their bulky bodies not fitting them for
life amid dense and tangled undergrowth. The deer of
temperate woods are, however, represented in the
tropical jungles, though perhaps the majority of the
tropical Asiatic deer tend to haunt relatively open
country. In Africa, where deer are absent, some of the

THE TROPICAL FOREST 107

antelopes have become modified for forest life. This is
especially true of the little duikerboks (Cephalophus),
whose slender bodies enable them to glide through the
undergrowth. The African harnessed antelopes (Trage-
laphus) also, and the African sable antelopes (Hippo-
tragus) are found in wooded country. In the African
tropical forest also is found the rare and little-known
okapi, a relative of the giraffe, but with stripes recall-
ing those of a zebra (Fig. 46).

In tropical Asia the sambar and its allies are the chief
deer of forests, but the muntjacs (Cervulus) also haunt
wooded country, choosing upland regions. In the
dense jungles of South-East Asia occur various members
of the genus Tragulus, including small, almost rodent-
like animals, of primitive structure, called chevrotains.
They are the smallest of living ungulates, being only '
about a foot in height, and their slender legs and bodies
enable them to glide through the jungle, and thus to
persist despite their generalized structure. An allied
form, the water chevrotain, occurs on the west coast
of Africa, where it inhabits swampy regions.

In a densely forested region, rivers or swamps, with
their special types of vegetation, obviously afford
advantages to animals not fitted for arboreal life
proper. We find, therefore, that such parts of the forest
have special animals. The wild pigs of India and the
adjacent regions, the bush pigs of Africa, are examples
of forms which haunt the damper parts of forests.
Africa has also its hideous wart-hogs (Phacochoerus)
while the islands of Celebes and Buru lodge in their
forest the curious babirusa with its enormously elon-
gated tusks. America has no true pigs, but the slender-
limbed peccaries replace them, and are similarly
inhabitants of the swampy parts of forests. All these

108 THE FAUNA OF

pig-like animals differ from those ungulates which live
on hard ground in the greater number of functional
toes, in the way in which these toes spread out to pre-
vent the animal sinking in the mud, and in the absence
of the elaborate arrangement which enables a sheep
or an antelope to swallow large quantities of partially
masticated vegetable food, which is regurgitated and
thoroughly masticated later, when the animals feel
themselves relatively safe from pursuit.

Among the odd-toed ungulates the tapirs, both the
Malayan form and those occurring in South America,
are forest animals, haunting much the same regions as
do the pigs. The great enemy of the American forms is
the jaguar, and the fact that when attacked the animals
endeavour to reach thick cover suggests that the forest
habitat is partially determined by the need for protec-
tion. As in the case of the pigs it is probably partially
determined also by the need for finding relatively soft
and succulent food, for the teeth of pigs and tapirs
have not the elaboration of structure found in the more
highly differentiated ungulates.

The rhinoceroses show somewhat different conditions
in regard to habitat, for while some haunt more or less
open plains, others, like the Javan and Sumatran
rhinoceroses of South-East Asia and the common
rhinoceros of Africa, are found in forested country, and
subsist largely on leaves and twigs. Horses are not
forest animals. The elephants, on the other hand, both
Indian and African, are found in dense jungles, but in
both cases this seems to be largely on account of the
animals’ great intolerance of heat, for they emerge
into the more open regions both at night and during
the cooler parts of the day.

Among the rodents, the squirrels and flying squirrels,

(S7OM *M Aq aangord ay} wos) “seolIoULY oY} 04 pouguoo st vund eyy, “uorye10[00
TaLofIuNn sMoys anpe oy} ‘peyjods ore Sunod oY} opty 9ey eI0N ‘sqnQ pue euNg ‘9g “SLT

Fic. 27. The African Elephant.
(from a specimen in the Royal Scottish Museum.)

THE TROPICAL FOREST 109

already described in temperate forests, are also found
in tropical ones, the squirrels reaching their maximum
number in the Malayan region. India and the adjacent
regions have large flying squirrels belonging to the

Fic. 28. Elephant in teak forest, Upper Burma. The position
brings out the peculiarities of the hind-limbs, which are very different
from those of cow or horse. Note especially that the long, straight
thigh is not bound down by skin as in the horse, and the true knee
appears to be very low down. (Photo by Col. Couchman.)

genus Pteromys, but Africa has forms which belong to
quite a different family (Anomaluridae). The Latin
name is given on account of the peculiar tail, which
has overlapping horny scales on its under surface near
the base, stated to be used in obtaining a hold on the
bark when the animals are climbing trees, and thus

110 THE FAUNA OF

offering some analogy to the stiff tail of the wood-
pecker (p. 47), which has the same function.

We have already spoken of the Canadian porcupine
(p. 43) in the taiga. This is represented in the
tropical forests of Mexico and Brazil by the tree-
porcupines of the genus Syntheres, which present
some interesting adaptations to arboreal life. The tail
is markedly prehensile, and the hind-foot is so inserted
as to make it easy for the animals to grasp the branches.
The great toe is absent, but its place is taken function-
ally by a fleshy lobe which can be bent inwards so
as to be partially opposed to the toes. The animals
are stated not to drink, and probably do not descend
voluntarily from the trees (see Fig. 30).

We have already repeatedly emphasized the fact
that ungulates in America are few, and that their place
in nature is taken there by rodents; there is, therefore,
nothing surprising in finding that the rodents called
agutis (Dasyprocta) haunt the forests of Brazil and
elsewhere, and have become adapted for life in the
dense jungle. With their slender limbs and bodies
they show a close, though entirely superficial, resem-
blance to the chevrotains of South-East Asia. The paca
(Coelogenys) has similar habits.

With the Edentates we come to a primitive order,
whose members have kept their hold in a world which
has grown beyond them either by retiring to the depths
of the tropical forests, or by acquiring burrowing and
concealed habits. In the forests of the Amazon occur
the sloths, belonging to the genus Bradypus and
Choloepus, perhaps the most truly arboreal of all
mammals, for they are quite helpless if removed from
the trees among which they pass their whole lives.
The fact that they possess practically no adaptation

SEAS ee ris =

Fig. 29. An African Flying Squirrel (Anomalurus). The characteristic
scales on the under-surface of the tail cannot be seen in this specimen,
but the bony structure projecting from the elbow, which helps to support
the parachute, is well shown. With this figure should be compared
that of the Flying Phalanger (Fig. 33). Though not nearly related the
animals show a superficial resemblance to one another. (Froma specimen

in the Royal Scottish Museum).

(unasn pr yssnoog phogy a2 U2 usur00ds D W027) “YIOS

(‘wnasnpr ysimoog pohoy 9y2 Us uauweds p wo.s7)
I oLstIojowIeyO oY} PUB [TB> epsueyerd

‘groqowreqo peeroqie qYOq ‘e0} yeers opqesoddo ey} pus quoiy yuepo
[req opsueyerd oy} ejony “umssodg womh oy, “18 “OWT oy ejony ‘oudnoz0g eoty, UeTZeIg “OF “OL

THE TROPICAL FOREST 111

to permit them to pass from one tree to another shows
that their natural habitat is the dense forests where
tree is bound to tree by lianes.

Among the special features we may note the nature
of the coat, the individual hairs having a fluted surface,
on which algae lodge, and so give the animals the
appearance of a lichen-covered branch. The fingers
are converted into mere hooks, the tail is a stump,
a common feature in such arboreal animals as have
not prehensile tails, or do not require to use the tail
as a balancing organ after the fashion of squirrels and
tree-shrews (cf. bears, anthropoid apes, &c.). We have
already spoken of the elongation of the forelimbs in the
sloths (p. 95). The food consists of leaves and fruits and
the animals do not drink. The usual position is hang-
ing back downwards from the branches of trees, and the
animals sleep rolled up in a ball with the head tucked
between the arms. Like many helpless forms they are
active only at night (see Fig. 32).

The South American ant-eaters are also forest animals,
the large form called Myrmecophaga jubata being strictly
terrestrial, while the little two-toed ant-eater (Cyclo-
turus didactylus) is arboreal, and except that it has
a prehensile tail and lives upon insects, has a curious
adaptive resemblance to a sloth, both in the struc-
ture of its limbs and in appearance. The hairy ant-
eaters of South America are replaced in India and
Africa by the scaly ant-eaters or pangolins (Manis), but
though some of these are partially arboreal, most are
found in rocky country, where they burrow in the
ground.

Not a few of the marsupials have acquired arboreal
habits. Thus the tropical forests of New Guinea and
Queensland are inhabited by tree kangaroos, which have

112 THE FAUNA OF

apparently recently taken to the trees, for they are
stated to be slow and clumsy climbers. In other
kangaroos the forelimbs are very short, in these forms
they are nearly as long as the hind ones, and the tail,
though not prehensile, is said to be used in climbing.
Much more definitely arboreal are the phalangers,
closely related animals, but as well fitted for life in the
trees as most kangaroos are for life on the ground. In
the forests of Western Australia occurs the little Tar-
sipes, a shrew-like animal with a prehensile tail, which
sucks honey from flowers with its worm-like tongue,
and presumably, like humming birds, obtains insects
also in this way. In North Australia and in the islands
of the Indo-Malayan region is found the spotted cuscus
(Phalanger maculatus). As in the other members of
the phalanger family the great toe can be opposed to
the others, and the long tail is very markedly pre-
hensile. Cuscuses feed upon leaves and fruit, but like
so many of the forest animals will also take birds and
other small creatures. There are several species, mostly
about the size of a cat. The feet are naked beneath.
Just as we have flying squirrels among the rodents
and flying shrews .among insectivores, so among the
phalangers (Fig. 33) we have also flying types. The
great flying phalanger (Petauroides volans) of Queens-
land and Victoria has both a flying membrane and
a prehensile tail ; the squirrel flying phalanger or sugar-
squirrel (Petaurus sciureus) has a bushy squirrel-like
tail, not prehensile, and a well-developed parachute.
It occurs in Eastern Australia. The pygmy flying
phalanger (Acrobates pygmaea), in which the total length
of head and body does not exceed about 24 inches,
has a long tail with the hairs arranged at the sides to
imitate the vane of a feather, and a slightly developed

(‘wnasnyg ysujoog pohoy ay) us uauroads v wou.,7) -ueuoods Zuray] oy} Ur weed ‘yeoo ABBeys ‘osreoo on} [req 4104s
oy} ‘squmt]-e107 Suoy o4g ‘poyeredes Sureq jo ojqedeout ‘sq1Srp o]-HOoy oy. O}0N “(sndAppsg) yAOTY pooy-ooy OUL “SE “OLT

Spar

i MANGE ATLA 9

Fic. 33. The Lesser Flying Phalanger of Eastern

Australia. (From a specimen in the Royal Scottish
Museum.)

THE TROPICAL FOREST » ae

parachute. It is stated to show marvellous agility in
leaping from branch to branch, and is doubtless aided
by its feather-like tail.

Another member of the phalanger family, the koala
or native bear (Phascolarctos cinereus) is very different
from its allies in appearance, and shows some curious
adaptations. It is sluggish in its movements, and, as
often happens in such cases, the tail has disappeared
(cf. bear, sloth, &c.). In the hand two of the fingers
can be opposed to the other three, the condition which
occurs in the chameleon. On the foot, as in other
phalangers, there are five toes, the great toe being
opposable to the others. The animal feeds chiefly
upon leaves, but is also stated to descend to the ground
in order to dig for roots. Curiously enough, it possesses
cheek pouches like an Old World monkey.

The last of the marsupials to be considered are the
opossums of South America, small insect-eating animals
with opposable great toes and prehensile tails, which,
though chiefly South American, are represented in the
United States by the widely distributed common
opossum (Didelphys marsupialis). They must be added
to the very considerable list of peculiar arboreal mam-
mals possessed by South America, for they do not occur
outside of the American continent. The majority
spend their lives in trees, though some spread to the
pampas of Argentine, and they take the place in nature
elsewhere occupied by insectivores, which are absent
from the South American continent (see Fig. 31).

Birds are so abundant in the tropical forests that we
can only pick out for special remark a few which illus-
trate points of interest. In spite of the gloom of the
jungle brilliantly coloured forms are common, a fact
well illustrated by the gorgeous birds of paradise, the

1404 H

114 THE}JFAUNA OF

most brightly coloured of animals, which haunt the
forests of New Guinea and the neighbouring regions.
They spend much of their time in the trees, to whose
branches they cling easily with their strong feet.

Among Picarian birds a whole division possesses the
climbing foot, that is one in which the first and fourth
toes are directed backwards, while the second and third
point forwards, and almost all these are arboreal birds.
They include the widely distributed woodpeckers, found
in tropical no less than in temperate forests, the jaca-
mars, handsome birds found in the South American
forests, and the toucans with enormously developed bills
and bright colouring. The purely frugivorous plantain-
eaters of Africa differ in that the fourth toe can be
turned backwards or forwards at will, though the birds
seem to dwell entirely in the trees. The climbing foot
is however best developed in the parrots, which are for
the most part thoroughly arboreal. The legs are short,
and the short, strong, hooked bill, of which the upper
portion is hinged as well as the lower, is used in climb-
ing as well as in cracking hard shells and kernels.
Parrots are chiefly tropical, but they extend much
further to the south of the tropic of Capricorn
than to the north of that of Cancer. As is well
known, their colouring is often very striking, green
predominating, as it often does among arboreal, fruit-
eating birds.

Of the parrots, the crested cockatoos are Australian,
and the gorgeously coloured macaws with their long
tails South American. The little love-birds are African,
as is also the familiar grey parrot with a red tail. The
parakeets are found in Africa, India, and Australia.
Of almost all parrots it may be said, as of the fruit-
eating mammals, that they show a marked preference

THE TROPICAL FOREST 115

for robbing man when they can, rather than limiting
themselves to wild products.

Not a few of the pigeons are arboreal. We may
mention especially the fruit-pigeons, found in South-
East Asia, and extending southwards to Australia.
The bill is distensible at the base to permit the birds
to swallow large fruits whole.

Among the mammals we found that in addition to
the truly arboreal forms there were a number like the
chevrotains and agutis, which took advantage of the
shelter afforded by the dense forest, through which
their slender bodies permitted them to force their way.
Quite similar conditions occur among the game birds,
where various members of the pheasant family live
in dense thickets, through which their wedge-shaped
bodies enable them to travel easily. Such birds fly with
considerable reluctance, preferring to trust to the thick
cover unless danger approaches too closely. Among
such forms we may mention the wood-partridges of the
Malay region, the spur-fowl of India and Ceylon, the
jungle-fowl, the guinea-fowl of West Africa, and so
forth.

Very remarkable is the condition presented by the
hoatzin (Opisthocomus) of South America, where the
young are hatched with claws on both the thumb and
index finger, which they use in climbing about trees,
after the fashion of a fruit-bat. The hoatzin is believed
by some ornithologists to be related to the game-birds,
but is a very primitive form.

Among reptiles the crocodiles and their allies haunt
the lakes and rivers of forest regions, where they. lie in
wait for the forest animals as they come down to drink.
A very curious arrangement of the breathing organs
enables them to drown their prey by holding the

H2

116 THE FAUNA OF

animals caught under water till death takes place,
the crocodile being meantime protected from choking
by the nature of the internal nostrils.

Of the lizards the flying forms (Draco) are interest-
ing. They occur in the Indo-Malayan countries, and
their parachute is a lateral fold of skin supported upon
the posterior ribs, which are greatly elongated. The
mechanism is thus entirely different from that seen in
any flying mammal. The body is greatly depressed,
and the ‘ wings’ when not in use are folded close to
the body.

The iguanas of tropical America and the West Indies
are also arboreal forms, but show no special adapta-
tions to this mode of life. They haunt especially trees
overhanging water, and are said to have the curious
habit of dropping into the water with a splash, some-
times from a great height, if alarmed. The most purely
arboreal of the lizards are, however, the chameleons,
specially characteristic of Africa, but also found in
India and Ceylon, and extending into Europe in Spain.
All have prehensile tails, and the hands and feet con-
stitute perfect grasping organs. The limbs are long
and slender, the forelegs especially so. The fingers and
toes form two groups, two being opposed to the other
three. On the fore-limb the first three fingers form an
inner bundle opposed to the fourth and fifth, which
are outside. In the foot the inner bundle is made up
by the first two toes only, the third, fourth, and fifth
forming the outer group. The animals are insecti-
vorous, and their power of colour-change is familiar
to all.

Of the snakes the tree boas of South America are
forest-haunting forms, with prehensile tails, and feed
chiefly upon birds. Like their allies the pythons and

THE TROPICAL FOREST 117

true boas, they are non-poisonous, crushing their prey
by the weight of their coils. The huge anaconda
(EZuneces murinus) of the Amazonian forests is semi-
aquatic, lurking in pools till its prey approaches the
water. It feeds largely on peccaries. The true boas
occur both in Madagascar and in tropical South America,
and inhabit the dense forests, through which they can
glide without difficulty. The beautifully coloured and
marked wood-snakes (Herpetodryas) of tropical America
represent there the tree snakes (Dendrophis) of the Old
World, all of which are arboreal in habit. The American
forms feed upon lizards and young birds. In India the
very agile whip-snakes (Dryophis) twine their slender
bodies round branches of trees, shooting out their heads
in order to capture their prey. Most other snakes, includ-
ing the most poisonous forms, are ground animals,
often inhabiting dry and sandy places. An exception
is the climbing tree-viper (Trimeresurus gramineus) of
India and Burma, with a markedly prehensile tail, and
a generally green tint, like the trees among which
it lives. ;

Of the amphibians we have already mentioned the
flying frogs of the genus Rhacophorus, which occur in
Madagascar, and also throughout South-East Asia.
A number of other frogs show more or less well-marked
arboreal habits, the commonest adaptation being the
development of some form of adhesive disks at the end
of the toes, to permit of the animals fixing themselves
to the branches of trees. This is seen in the family of
Dendrobatinae, whose members occur alike in Mada-
gascar and in tropical America. They are beautifully
and often vividly coloured, and some species at least
have poison glands in the skin, whose secretion is used
in South America as an arrow poison. Such an associa-

118 THE FAUNA OF

tion of bright colouring with poisonous or noxious
qualities is of common occurrence in animals.

The typical tree frogs of the family Hylidae are
widely distributed over the globe, but are absent from
Africa south of the Sahara, from Arabia, and from
India. They have suctorial disks at the ends of their
toes, are usually protectively coloured in shades of
green, the tints varying with the surroundings, and.
have glands upon the under surface whereby they can
absorb water from the damp leaves on which they rest,
and thus obviate the necessity for seeking ponds to
moisten the skin. Their breeding habits are often
peculiar. Other amphibians are not adapted for the
arboreal life.

In regard to fresh-water fish we may note one or two
peculiar forms, found in the streams of the tropical
forest, or on its seaward margin. The very peculiar
double-breathing fishes, which are animals furnished
both with lungs and gills, are represented in the tropical
streams of Queensland by Ceratodus, in the tropical
streams of Africa by Protopterus, and in those of
South America by Lepidosiren. None of these animals
voluntarily quit the water, but the fact that the two
last named can breathe when buried in the mud as well
as when swimming in water adapts them for life in
tropical regions, where there is often a well-marked
alternation of wet and dry seasons.

Even more curious are the habits of the mudskippers
(Periophthalmus), which haunt the mangrove swamps
and mud-flats on the shores of the Indian and Pacific
Oceans and off the coast of West Africa. These are
bony fish which are partially adapted for life on land.
The anterior fins are curiously modified, so that the fish
can climb about the supporting roots of the mangroves,

THE TROPICAL FOREST 119

and it can also leap and skip over the mud. The eyes
are very prominent, are furnished with distinct eyelids,
and are protrusible. The fish, which are very tadpole-
like in appearance, feed upon insects.

In India and the south-east of Asia generally, another
fish is found which can live on land, and voluntarily
leaves the water to do so. This is the climbing perch
(Anabas scandens), a freshwater form which travels
long distances over land by hitching its pectoral fins
round plants. The fish is also stated to be able to
climb trees. It has a special accessory breathing organ,
which enables it to breathe air when out of the water.

In regard to the invertebrates, a special feature of
the tropical forests is the great wealth of insects. That
this must be so will be clear when we reflect that
some of the forest-dwelling primates, the insectivores,
many of the marsupials, some carnivores, the eden-
tates, many reptiles and amphibians and so on, eat
insects, either solely or in combination with other types
of food. Further, the equal temperature allows the
insects to go on breeding all the year round, and thus
permits a great wealth of individuals to occur, while
the special conditions favour a great wealth of genera
and species. Apart from the size and beauty of the
tropical butterflies, the most impressive of the insects
are first the flies, which are often blood-suckers, and
then may, as in the case of mosquitoes and tsetse flies,
serve as transmittors of the tropical diseases most fatal
to man and beast, and second the ants, some of which
are extraordinarily destructive to plant life, while
others from their numbers and ferocity are the enemies
of all other animals. The driver ants of West Africa,
which are wandering forms, are stated to devour every
living creature in their path which cannot make its

120 FAUNA OF THE TROPICAL FOREST

escape, and will even attack human beings if from any
cause these are rendered incapable of defence or flight.

The termites, or so-called white ants, are as a rule
less obvious in the dense forest than in more open
country, but some species do occur there. Other
members of the same order form the beautiful stick
and leaf insects, predatory forms which often show
a marked resemblance to the vegetation among which
they live. As well as with insects the tropical forest
swarms with scorpions, spiders, including the large
bird-catching form of South America, and ticks, which,
like flies, serve for the transmission of deadly diseases.
As most of these forms do not, however, show special
adaptations to forest life, it is sufficient to merely
indicate their abundance, which has a marked effect
upon the life of other organisms in forest areas.

We may add that, just as a few fish in tropical
climates have taken advantage of the greater oppor-
tunities which the land offers as compared with the
sea to acquire partially terrestrial habits, so various
crabs have become partially adapted for life on land,
and wander about in those forest regions which are at
no very great distance from the sea. to which they
return for breeding purposes.

Rererences. In regard to the animals of the African forests, Sir
H. H. Johnson’s works give very full details, with many figures; see
British Central Africa (London, 1897); The Uganda Protectorate (2 vols.,
London, 1902); Liberia (2 vols., London, 1906). Lydeker’s Game
Animals of Africa (London, 1908) should also be consulted. Wallace’s
Malay Archipelago (London, 1869) gives a fascinating account of the
animals of that region, and the same author’s Tropical Nature (London,
1878) may also be mentioned. Bates’s Naturalist on the Amazons (5th ed.,
London, 1884), and Belt’s Naturalist in Nicaragua (2nd ed., London,
1888), give good accounts of South and Central American animals. For
details as to the animals, reference should again be made to the natural
histories already mentioned.

CHAPTER VI

TROPICAL SAVANAS AND DESERTS

In the tropics the steppes of temperate latitudes are
represented by what are called savanas, which are
regions where grass occurs, mingled with arborescent
or shrubby vegetation. Sometimes we have in these
regions scattered trees only; sometimes clumps of
trees occur, separated by regions without trees, giving
a characteristic park-like appearance to the landscape.
Elsewhere the trees follow the courses of the streams,
forming the so-called gallery forests. Such savanas,/
occupy large stretches in the tropical regions of Africa,
South America, and Australia, and have a highly
developed and characteristic fauna.

The typical savana climate is tropical. The rain
comes in the hotter season of the year, and there is
a well-marked dry season. Where the rain is deficient
the savana passes into desert, as we see in passing
northwards from the Sudan to the Sahara, or inland
from the Australian bush to the central desert. On-the
other hand, where the rainfall is abundant the trees
increase in number, and the savana passes into the
tropical rain-forest, through the intervention of the \j
savana wood, with its numerous trees. The period of
greatest vegetative activity accompanies or follows the
rainy season, and the dry season may witness an almost
complete cessation of activity. This, like the similar
conditions in the steppes, involves a certain amount of
migration among the inhabitants of the region. Further,

122 TROPICAL SAVANAS

owing to the great power of the sun, bush fires are
frequent, and often have devastating effects.

The type of vegetation naturally varies in the
different regions in which savanas occur. In Africa
the mighty baobab (Adansonia digitata) is conspicuous
among the trees. It stores water in its thick trunk,

Fic. 34. Savana, with scattered acacias, on the margin of the desert
in Kordofan. (Photo by Capt. Lloyd.)

and becomes leafless in the dry season. Other important
trees are thorny acacias, whose foliage is relished by the
giraffe, various species of cactus-like Euphorbia, palms,
and so forth. Various kinds of grasses occur, and often
grow to a great height, forming dense thickets, which
offer much shelter, while thorny and xerophilous plants
are of course frequent. In the savanas of the other

AND DESERTS 123

continents, though the species differ, the general facies
is the same—tall grasses, thorny acacias, cactuses or
cactus-like forms, and generally plants presenting
special features which protect them against periodical
drought, while allowing them to take advantage of the
periodic tropical downfalls.

Of the great savana regions of the world the African
is especially rich in ungulates, notably in antelopes,
the Australian in herbivorous marsupials, notably
kangaroos and wallabies, while the savanas of South
America, a country which has witnessed an extra-
ordinary destruction of mammalian types in recent
geological time, were relatively poor in mammalia till
the advent of European man.

One of the special climatic features of the savana
proper is the periodic abundance of rain. As we pass
from the savana to the desert the dry season increases
in length, and the rain becomes less, and more un-
certain. Vegetation also gradually diminishes as the
conditions become less favourable, and the fauna
becomes impoverished, the mammalian fauna diminish-
’ ing first. In the desert proper few organisms can live,
but the less unfavourable regions carry a reduced
savana fauna, especially of insects and reptiles. The
Kalahari, indeed, is said to serve as a reservoir for such
insects as locusts, which breed there until starvation
forces the swarms to sweep outwards to the better
watered lands on the margin of the desert.

No hard and fast line separates savana regions from
steppes, this being especially true in South America,
where the treeless ‘pampas’ or steppes of the Argentine
south of lat. 32° S. pass into a savana region further
north. It should be noted, however, that in addition
to the characters already given, typical savana regions

124 TROPICAL SAVANAS

are distinguished from the typical steppe region already
described in the absence of a severe winter.

Proceeding now to consider the animals of the various
savana regions in systematic order, we may note that
as a general rule the primates do not occur in such
regions, being, as we have seen, mostly fitted by nature
for life in the tropical forest. To this rule, however,
the baboons form an exception, for, as we have already
seen, they are quadrupeds, living in open rocky country.
The Arabian baboon (Papio hamadryas) occurs abun-
dantly in the Sudan, where it feeds upon fruits, shoots,
buds, and seeds, and finds the necessary shelter in
rocky ground. In Angola another species (P. anubis)
inhabits very dry country, and feeds largely upon that
strange coniferous plant known as Welwitschia.

Among the insectivores we may note especially the
jumping shrews, which are confined to Africa, and
show a curiously close resemblance to the jerboas among
rodents. The jumping or elephant shrews have greatly
elongated hind-legs, and move in a series of leaps. They
have also curiously elongated snouts, which they use
in routing about for the insects on which they feed.
The type genus is Macroscelides, and the animals are
widely distributed in Africa, being mainly but not
exclusively tropical. The elongation of the hind-limbs
is interesting, for it is frequent in steppe and savana
animals, and has several obvious advantages which have
been already discussed (see p. 64).

Of the carnivores, the large cat-like forms, as we
have already had occasion to remark, show no very
special adaptation to one habitat. The uniform colour-
ing of the lion, however, suggests that it was once
a desert animal. In Africa it is abundant in the Kala-
hari desert, but also extends into savana and steppe

AND DESERTS 125

regions, its distribution being probably determined by
the double need of avoiding the proximity of its per-
sistent enemy, man, and of following its natural prey,
the larger ungulates. Of the smaller cat-like forms, the
mongooses are ground animals, usually haunting open
country, and avoiding dense forests. An allied form,
the meerkat (Suricata tetradactyla) haunts sandy ground
in Cape Colony, where the savanas pass into steppes.
Like most of the smaller mammalia of open country
this animal is a burrower, making deep holes in
the sandy veldt with its sharp foreclaws. Another
adaptation to life in this region is shown by the fact
that it feeds chiefly upon the bulbs and roots of
the veldt plants, which store water in their under-
ground parts during the time of drought. Another
steppe or savana form of wider distribution is the
aard-wolf (Proteles cristatus), a burrowing animal, with
weak teeth, which seems to feed largely upon termites,
and inhabits open country throughout the southern
half of Africa, with a northern limit apparently in
Somaliland, where it has been once found. The hyaenas
are also inhabitants of savana country, the spotted
hyaena of Africa being often found in the same districts
as the lion, upon which it is partly dependent, in that
it obtains the remainder of the ‘ kill’, when the nobler
animal has satisfied its hunger.

As a rule the members of the dog alliance inhabit
open country, as the common habit of hunting in packs
suggests. A considerable number occur in the tropics,
and may therefore be included in the fauna of deserts
and savanas. We may note especially the African
jackals, which hunt in packs and feed upon rodents, or
the smaller ungulates, or sick or wounded members of
the larger species. An interesting desert species is the

126 TROPICAL SAVANAS

long-eared fox, or fennec, which inhabits the Sahara
desert, and is called Canis zerda. This little animal
has very large ears and great acuteness of hearing ; it
is tawny-coloured like the desert sand, and is burrowing
and nocturnal in habits. It burrows with great rapidity,
being said to appear to sink through the sand. At.dusk
it becomes active, and sets forth in search of insects,
lizards, small birds, rodents, or even fruit if obtainable.
The animal is partially social, the burrows being made
in company. A very much fiercer animal is the Cape
hunting dog (Lycaon pictus), which has a curious and
unexplained resemblance to the spotted hyaena. It
is widely spread throughout Africa, where it inhabits
open country, preying upon the ungulates, which are
borne down by sheer weight of numbers. In South
Africa the flocks of the white man form a favourite
source of food. Though the animals live in holes, the
young being born underground, yet on an alarm it
seeks safety in flight rather than in the burrow like the
fennec. It would appear that the dogs have little
or no burrowing power themselves, their holes being
either natural or obtained by ejecting the original
occupant. So swift are these dogs that they are said
to be able to overtake the swiftest antelope. Not a few
other dog-like animals haunt the African savanas, but
these may serve as types.

Bears, as we have already seen, are absent from
Africa south of the Atlas, and one species only occurs
in South America, and that in the Andes. As these
two countries have the best-developed savanas, it is
clear that the animals are unfitted for life in such
regions. They are indeed absent alike from temperate
steppe and tropical savana and desert. The allies of
the bears, such as the raccoons and coatis, are also

AND DESERTS 127

absent from savanas, and the weasel group is not well
represented there. Among the few weasel-like forms
we may note the South African weasel (Poecilogale) and
the Cape polecat (Ictonyx). The latter has a somewhat
wide distribution in Africa, and differs from martens
and polecats in that it is unable to climb. It has
a peculiar resemblance to the American skunks, and
like them is protected from possible enemies by its very
offensive odour. The animals feed upon small mammals
and birds, lizards, and frogs, and are found in rocky
districts.

It is when we come to the ungulates, however, that
we find the most specialized and abundant of the
animals of the African savana. In the reedy swamps
throughout the greater part of the continent occurs the
so-called Cape buffalo (Bos caffer), which is replaced in
the regions richer in trees by the short-horned buffalo
(Bos pumilus). Both live in herds, and do not volun-
tarily quit the vicinity of water, in which they love to
bathe. The young are born in the warmer season, and
there is never more than one calf at a birth. It is
hidden by the mother among long grass, and for about
ten days after its birth the mother separates from the
herd, and remains within easy reach of her young,
which she visits at short intervals. This habit speaks to
life in regions where cover is always obtainable, and is in
striking distinction to the habits of the mountain ungu-
lates (cf. p. 75), where the young must be able to
travel with the herd very shortly after birth.

As we have repeatedly stated, it is the antelopes
which characterize specially the savanas of Africa,
the diversity of species being as remarkable as the
number of individuals, till man interfered. In several
points of structure antelopes are more primitive than

128 TROPICAL SAVANAS

oxen or sheep and goats, and they appear to be
older.

Eland are the largest African antelopes, and are
distinguished by the presence of horns in both sexes.
They were formerly common over the whole of
Eastern and Southern Africa. Typical savana animals,
they prefer open plains with scattered timber,
but extend into the desert on the one hand and into
the open savana wood on the other. Like not a few
tropical animals they are intolerant of the hot sun,
and where possible spend the hours of greatest heat in
the shelter of woods, moving back to the grassy plains
to feed and drink. Where possible they drink daily,
but unlike the buffaloes they can tolerate thirst, and in
the Kalahari desert apparently obtain sufficient water
by eating water-melons and similar succulent fruits or
plants. Like most ungulates they live in herds, and the
student of botany will notice that the fodder plants
upon which these animals depend (grasses, &c.) are
social also, while in the tropical forests, where social
animals are rare, social plants are also infrequent
(cf. p. 96).

Kudu (Strepsiceros) are distinguished from eland by
the absence of horns in the female, as well as by other
characters. They haunt thickets and country covered
with bush, and occur in small parties, usually on very
rough ground. Their speed is not very great. The
preference for bush or thicket-covered country is even
more marked in the harnessed antelopes (Tragelaphus),
which we have already mentioned as being found even
in the tropical forest. On the other hand, the addax
from the deserts of North Africa and Arabia, and the
gemsbok (Oryx gazella) from the deserts of South-
Western Africa, are examples of forms which inhabit

AND DESERTS 129

very arid country, and appear capable of going for long
periods without tasting water. The horns are very
powerful, and the gemsbok is stated to be able to beat
off the lion by means of them, but not to be swift,
perhaps because it had few enemies till man appeared
on the scene. In addition to the gemsbok, other
species of the genus Oryx occur in different parts of
Africa, all being inhabitants of open country.

The gazelle group, largely represented in the more
arid regions of Asia, North Africa, &c., has a South
African member in the springbok, once enormously
numerous on the margin of the Kalahari desert. Other
African species also occur. An allied form of interest-
ing habits is Waller’s gazelle (Lithocranius walleri) from
East. Africa, which inhabits rocky districts, where
thorn-jungle occurs, and has a remarkably long neck,
giving it a resemblance to a miniature giraffe. It feeds
chiefly upon twigs and leaves, and is stated to rest its
forefeet against the trees, in order to reach the foliage.

Among other forms we may notice the pala antelope
(Aepyceros melampus), found among the acacia scrub
of Southern and South-Eastern Africa, and remarkable
for its great swiftness, and for its power of taking long
leaps. It prefers the proximity of water. In swampy
regions where reeds are plentiful the reitbok and water-
buck antelopes occur, the latter seeking the water or
swamps when pursued. On the other hand, hilly and
mountainous districts are inhabited by the active
rhebok, which resembles a chamois. Of the little klip-
springer we have already spoken (p. 84). The steinboks,
also small animals, occur in open country or savana
wood, but avoid hilly country. They are remarkable
for their speed.and their close resemblance in colour to
the ground.

1404 I

130 TROPICAL SAVANAS

The last group of African antelopes includes the
large wildebeest or gnu, the hartebeests, and the bless-
bok. The wildebeests (Connochoetes) are ungainly
looking animals, with some resemblance to oxen, which
inhabit open country in South and East Africa. The
genus Bubalis includes the somewhat stag-like harte-
beest, the blessbok, and the bontebok. All the three
mentioned are exclusively African, though one member
of the genus extends into Syria, and frequent grassy
plains, especially those which have so great an exten-
sion in Southern Africa.

Generally, we may say that very many of the African
antelopes are typical savana animals, but some, like
the gemsbok, extend into the desert proper, a few, like
the duikerboks, into the tropical forests, while not a few
extend their range up the mountains, the klipspringer
being a typical example, and some, like the water-buck,
frequent the swamps, and seek safety there. Those
which inhabit open plains find security in their numbers,
the strength and powerful horns of the bucks, and their
swiftness. Less powerful and slower forms must haunt
country which offers some form of shelter, as e. g. rocky
regions, thorny jungle, or swampy districts.

The resistance which these wild forms offer to the
diseases carried by tsetse-flies and ticks must have been
a factor in their persistence, for these diseases form the
greatest obstacles to the introduction of domesticated
ungulates into the great plains of Southern Africa.
The wild antelopes (cf. negro children and malaria)
are apparently tolerant of the parasites of the various
diseases, which affect them but little. They thus serve
as reservoirs of infection, by means of which introduced
forms may be infected. The virtual absence of wild
ungulates in South America when it was colonized from

(wnasnyy ystnoog poRosy
ay} ws uaunoads D WO) “BolITy Jo sodopoyue oy} Jo ouo “(soze0qoOMU|) NUY JO YSoSqePTIM PI!S4-OF1TM OUT, “CS ‘HLT

ticel

Fie. 36. The Giraffe drinking, showing the great difficulty which the animal
has in reaching the ground, in spite of the length of the neck. (From a
specimen in the Royal Scottish Museum.)

AND DESERTS 131

Europe meant the absence of such reservoirs of disease,
and accounts for the greater ease with which domesti-
cated ungulates were introduced there. It has been
suggested as a reason for the disappearance of so many
ungulates from South America—for they were once
abundant there—that they were killed off by some
parasitic disease. If this were so, then with the dis-
appearance of the hosts the parasites must have died
also, leaving the field clear for reintroductions later.
In Africa we have to notice that the tolerance of minute
blood parasites—a tolerance doubtless originally ac-
quired at the price of a fearful death-rate—is a means
of protection of the existing forms against the intru-
sion of new and competing forms. The resistance to
the trypanosomes carried by the tsetse-fly has even to
some extent protected the antelopes from their great
enemy man, for European hunters are constrained to
avoid regions much infested by the tsetse on account
of the difficulties of transport.

Another very beautiful ungulate which frequents all
the open parts of Africa south of the Sahara is the
giraffe, which shows a peculiar adaptation to life in the
savana. The great elongation of the neck and of the
forelimbs enables it to browse upon the leaves of
the acacias and other trees of the scrub. Despite the
length of the neck, however, drinking presents great
difficulties, the animal being compelled to straddle its
legs apart before it can reach to the ground. Grazing
is similarly difficult, and is rarely practised. The giraffe
inhabits arid country, notably the Kalahari desert, and,
generally, sandy plains where the scrub on which it
feeds occurs. It can apparently go many months
without drinking. The animals go about in herds, and
are capable of great speed. Only a single young one is

I2

132 TROPICAL SAVANAS

produced at a birth, and when three days old it can trot
by the side of its mother, who protects it from the
attacks of carnivores by kicks with her powerful legs.
The large eyes and the long neck give the giraffe a
very wide range of view, very necessary in an animal
which frequents open country.

Deer are usually absent from open plains, and are
entirely absent, as we have seen, from Africa south of
the Atlas. But in South America, where antelopes are
totally absent, the deer extend their range to the plains
and swamps. Thus the pampas of the Argentine and
Paraguay, together with similar regions further south,
are inhabited by the pampas deer (Cariacus campestris),
which finds the necessary shelter among the long
pampas grass. This deer occurs in pairs or small
parties, and the female is particularly ingenious in pro-
tecting her fawn. The latter makes off through the
grass on an alarm, and then cowers down, while the
mother takes herself off in another direction in a slow
and limping manner.

We have spoken of the camels of the steppes of the
Old World, and have seen also that their allies, the
llamas of the New, are intolerant of great heat, so that
the tropical savanas, whether in Africa or in America,
have no members of this family.

Of the odd-toed ungulates the rhinoceroses, as already
mentioned, extend into the plains, this being especially
true of Burchell’s rhinoceros, a grass-eating form. Of
the striped horses of Africa Burchell’s zebra and the
quagga are (or were) both inhabitants of the open
grass-covered plains, but both are chiefly extra-tropical,
inhabiting the vast plains of the extreme south of
Africa. The African wild ass is similarly an inhabitant
of the arid regions of North Africa. While Burchell’s

AND DESERTS 133

zebra and the quagga formerly occurred in vast herds,
the wild ass, an inhabitant of a region where food is
always scanty, occurs in smaller parties. Its sober tint
harmonizes with the uniform colour of the desert, and
the fact that the domesticated form will eat dry and
prickly food points back to the days when its ancestors
contented themselves with the thorny herbs of the
desert. The stripes of the zebra and quagga apparently
harmonize with the light and shade playing through
the open bush country of their native haunts. It will
be noted that the presence of the various forms of wild
horses on the plains to the north and south of Africa,
and their limitation chiefly to elevated regions in the
tropics (Grévy’s zebra), speaks to a relative intolerance of
heat, a not uncommon feature in the larger ungulates.

As to the rodents, we note that in Africa those of
the savanas are relatively unimportant, while in South
America, with the paucity of ungulates, there are many
large and important rodents.

We have already spoken of the jerboas, which extend
from the steppes of Asia into the deserts of North
Africa, and have noted their special adaptations. In
the southern half of the continent the place of the
jerboas is taken by the so-called jumping hare (Pedetes
caffer), very common at the Cape, but extending north-
wards into Angola and Mozambique. While related to
the jerboas the animal resembles the common hare in size
and colour, but it has a long hairy tail, and elongated
hind-legs. It feeds upon all-fours, but escapes from its
enemies by taking long bounds, stated sometimes to
cover as much as thirty feet. The animals are bur-
rowers and nocturnal, and frequent arid districts.
True hares also occur in Africa, but offer no special
features.

134 TROPICAL SAVANAS

Among the various mole-like forms of Africa mention
may be made of the little sand-rats (Heterocephalus)
of the sandy deserts of Somaliland. These little animals
—which are no bigger than a mouse—are nearly naked,
perhaps as an adaptation to life in sandy soil. They
resemble moles in having no external ears, and in their
almost functionless eyes.

. Of the rodents of South America the majority have a

wide distribution, ranging from the true savanas south-
ward to the pampas of the Argentine and the plains of
Patagonia. The viscacha (Lagostomus trichodactylus),
according to Mr. W. H. Hudson the most typical
inhabitant of the pampas, extends beyond their limits,
though scarcely into the savanas proper. On the
other hand the mole-like tucotucos (Ctenomys) are
burrowing animals, which occupy sandy regions from
Brazil and Bolivia to the extreme south. The coypu
(Myopotamus), an aquatic form haunting the banks of
rivers and lakes, has an even wider distribution. Many
species of cavy also occur, some being partial to moist
situations, while others dwell in rocky regions. The
largest living rodent, the capivara or capybara (Hydro-
choerus capivara) has a very wide distribution, and is
aquatic in its habits, frequenting the margins of lakes
and rivers, but also feeding on the open plains.

Of the South American Edentates most of the arma-
dillos inhabit open country within the tropics, though
some extend into extra-tropical regions. Most prefer
arid regions, and feed on insects and carrion, together
with some vegetable matter. The tiny pichiciago or
fairy armadillo (Chlamydophorus): is found in sandy
regions in the western part of the Argentine where the
vegetation consists of cactuses and thorny brushwood.
It burrows with great rapidity, and is very sensitive

Fie. 37. Jerboa (Dipus) from Central Asia.
(From a specimen in the Royal Scottish Museum.)

Fic. 38. The Egyptian Jerboa (Dipus). The animal is of a greyish-
sandy colour which renders it very inconspicuous in the desert regions
in which it lives. (From a specimen in the Royal Scottish Museum.)

AND DESERTS 135

to cold or wet. Another species is found in Bolivia.
The members of the type genus Dasypus are widely
distributed, being especially common in the Argentine.
In addition to the protective covering of bony plates,
they find security from their enemies in their burrow-
ing habits, and in the fact that they are mostly noc-
turnal.

In Africa, where armadillos are completely absent,
another curious edentate, the aard-vark or ant-bear
(Orcyteropus) occurs in open regions, and is also of
burrowing nocturnal habits, though it has no armour
as the armadillos have. It seems to feed chiefly on
termites, the so-called white ants.

Of the marsupials the kangaroos and their immediate
allies are exclusively confined to the Australian region,
and there they formed the natural fauna of the savanas
till the advent of the European with his flocks and
herds. The kangaroos and wallabies, with the excep-
tion of the tree-kangaroos already mentioned, are
ground forms, feeding not only upon grass but also
upon shoots of bushes and shrubs, or even in the case
of the larger forms, on leaves, &c., of trees, which they
reach by standing in the upright position. The tail is
long and strong. It aids the animal in maintaining
the upright position, and also in running—that is, in
leaping. We have already noticed the essential points
of structure in the kangaroo; the elongation of the
hind-limbs, without corresponding elongation of either
forelimbs or neck, is an adaptive peculiarity which is
far from uncommon in animals of the open plains. In
the ungulates the other type of adaptation, that which
consists in the elongation of both pairs of limbs, and
therefore necessarily of the neck to permit feeding to
take place, is the usual one. Just as among the rodents

136 TROPICAL SAVANAS

we have various stages in the elongation of the hind-
limbs, and therefore in the development of the power
of leaping, so among the kangaroos we have forms like
the rat-kangaroos (Potorous) in which the elongation
of the hind-limbs is much less marked, and which run
on all fours, and cannot deliver the powerful kicks which
are so important a part of, e.g., the giant kangaroo’s
means of defence. In the scrub of Queensland the five-
toed kangaroo (Hypsiprymnodon moschatus) occurs, and
in the structure and shortness of its hind-feet approaches
the phalangers, showing that the true kangaroos have
been evolved in response to the special conditions which
prevail on the open plains, just as the jerboas on the
Asiatic steppes have been evolved from ordinary
rodents, or as the long-legged antelopes and giraffe
have been evolved from the ancestral short-legged
ungulates, in response to similar conditions. The nature
of these special conditions has been already emphasized,
but we may repeat that if an animal of the open plains
cannot burrow, or defend itself passively in some way,
it must be swift to escape its enemies, it must have
a wide range of vision in order to note their advance,
it must have acute senses. We saw also in connexion
with the ungulates that there is a marked tendency to
reduction in the number of digits, to a fusion of limb
bones, and so on, in order to give the necessary strength
and rigidity for rapid movement. That the same need
produces the same results is obvious when we note that
the kangaroo has four toes on its hind-foot instead of
five, and that of those four two are rudimentary, likeghe
useless side toes in antelope or deer, though in the
kangaroo the rudimentary toes have apparently a sub-
sidiary use, for they are said to be employed in cleaning
the fur.

AND DESERTS 137

In one respect, however, the kangaroo shows a very
striking contrast to the ungulates. We have noted
that among the swift ungulates the tendency is for the
young to be born very fully developed, able after a very
short period to run by the side of the parent. The
consequent long period of gestation is dangerous to
the mother, but has apparently been justified in the
struggle for existence by the higher survival rate of
the young. Further, as we may note readily among
domesticated sheep, the young ungulate, when suck-
ling, is not allowed by the mother to satisfy its appetite
completely—it must never be so satiated as to be in-
capable of flight. In the case of the kangaroo con-
ditions are very different. The period of gestation is
very short, and the young one is born in a very im-
pertectly developed condition, incapable of sucking and
quite incapable of voluntary movement. It is placed
by the mother in the pouch with which she is provided,
and attached to the teat. Milk is then pumped into
its mouth, by the action of certain of the muscles of
the mother. Owing to this abundant food, growth is
rapid, and soon the young one is able to move and
leave the pouch. It however always returns to it
on an alarm, and is not able to care for itself
till it is some eight or nine months old. The mother
shows great devotion to her young, but if hard pressed
by dogs she will throw the young one out of the
pouch into the fern, thus relieving herself of a weight
which diminishes her speed. The action has been
variously interpreted. According to some authorities
the chances are that the dogs will neglect the young
one and continue to chase the mother, who, if she
escapes, will return to pick up her ‘joey’. Others
regard the action as equivalent to an abandonment of

138 TROPICAL SAVANAS

the young. In any case the fact that, with rare excep-
tions, marsupials have only succeeded in surviving
in the Australian region, suggests that the placental
method, that practised by the ungulates, has been
justified in the struggle for existence as against the
marsupial method, despite the greater risk to the
mother before birth and at birth in the former method.

Only one kangaroo is produced at a birth. This is
an interesting point, because a reduction in the number
of young is frequent in animals which must escape
with their young by flight when threatened, and do
not attempt to defend themselves nor to hide. A similar
tendency is well marked in the higher ungulates. But
it must not be supposed that the kangaroo is entirely
helpless. In the ungulates we find that, though the
primary purpose of the elongation of the limb is to
give swiftness, yet secondarily, e.g. in the horse, the
powerful hind legs can be used as weapons. Similarly
the kangaroo can use its long hind-legs to deliver heavy
blows, capable of killing a dog. Like so many denizens
of open country, the animals are markedly gregarious.

In regard to the other marsupials we need only note
that Australia contains various forms showing special
adaptations to open grassy plains or desert regions,
these adaptations exhibiting a curious convergent
resemblance to those which appear among the pla-
centals. For example, throughout the greater part of
Australia the open grassy plains are inhabited by the
so-called pig-footed bandicoot, a small animal with
long hind-legs, remarkable in showing an ungulate-like
reduction in the number of the toes. The fore-foot has
two functional toes, the hind only one; the animals
take both vegetable and animal food. The jerboas and
jumping mice of other desert regions are represented

AND DESERTS 139

in Australia by small marsupials of similar habit,
notably by Anéechinomys laniger, a little jerboa-like
animal with long hind-legs, a bushy tail, and the habit
of progressing by jumps. Again, the banded ant-eater
(Myrmecobius fasciatus) shows the long extensile tongue
common to ant-eating forms, and is found in sandy
regions where ant-hills occur.

Among the inhabitants of savana regions, mention
must be made of the running birds, which are specially
adapted to these districts. The adaptations are well
seen in the African ostrich (Struthio), distributed
throughout Africa wherever the open sandy regions
suitable to its habits occur, and found also in Syria,
Arabia, and Mesopotamia. The great length of the
legs and of the neck gives the animals a wide range of
vision (cf. giraffe), the length of the hind-legs and the
reduction of the number of toes to two gives them
speed (cf. again the giraffe). Like the large antelopes
with which the ostriches associate in the southern parts
of Africa, the ostrich is a social animal, the gregarious
habit, as in the ungulates, being associated with poly-
gamy. The males are bigger and stronger than the
females, and use their hind-legs as weapons when
attacked. The young are active almost immediately
after hatching, again recalling the precocious young of
the ungulates of the steppes and savanas. Like many
savana animals, ostriches can go without water for a pro-
longed period, but they drink freely if water is available.

In South America the place of the ostrich is taken
by the rhea, which has three toes instead of two, and
some other structural differences. The habits are some-
what similar. The birds are chiefly found on the
pampas, which correspond rather to steppe regions
than to savanas in the strict sense.

140 TROPICAL SAVANAS

In Australia and the neighbouring regions two other
types of running birds occur, these being the emus and
cassowaries, which resemble the rheas in having three
toes, but of these the cassowaries are forest birds,
rarely venturing out into the open. The emu, on the
other hand, is a savana bird, haunting the plains and
open country, where it eats a large amount of grass.
The remaining living member of the group, the kiwi of
New Zealand, like the cassowary, is an inhabitant of
wooded districts.

Another interesting group of birds which inhabit the
grassy plains of South America, though extending also
into forest regions, are the tinamus, which show certain
resemblances to ostriches, but possess the power of
flight. They strongly resemble game birds in outward
appearance, being popularly called partridges, but they
do not fly nearly so well as the game birds, and rise
with reluctance, preferring to skulk among the long
grass. The flight is swift, but cannot be kept up for
long at a time.

Among other birds characteristic of savana regions
mention may be made of the bustards, of which many
species occur in Africa, just as the common form occurs
in the steppe regions of Asia. The curious secretary
bird (Serpentarius secretarius) also, with its long legs, is
very characteristic of the African savanas. Though
it is capable of flying well and strongly, it prefers to
run along the ground, like the ostrich. Its structure
shows that it is allied to the birds of prey, and the fact
that the young are very helpless, and are for long
unable to run upon their elongated legs, suggests that
the ground-haunting habit is a recent acquisition. The
nest is a huge structure and is often placed in a mimosa
bush.

AND DESERTS 141

Fic. 39. The Cactus Wren (Campylorhynchus brunneicapillus), a
characteristic bird of the desert regions of California and Texas, where
it builds a complicated nest in cactus bushes. (Photo by the Biological
Survey, U.S.A.)

In South America the place of the secretary bird in
nature is taken by the seriemas, birds apparently.
related to the cranes, but showing a curiously close

142 TROPICAL SAVANAS

external resemblance to the secretary birds. The
Brazilian seriema (Cariama cristata) haunts open dis-
tricts with scattered bushes, and is a cursorial bird,
though it roosts in trees.

There is nothing very characteristic about the
reptiles of the savanas and hot deserts, but of the
numerous lizards which occur there one or two may be
named as showing interesting peculiarities. In the
sandy districts of Western and Southern Australia occurs
Moloch horridus, a lizard covered with spines and
tubercles. It lives upon ants, and has the curious
power of being able to absorb water through its rough
skin. This is presumably an adaptation to permit the
animal to avail itself of the rare showers which fall in
the desert regions where it lives.

Somewhat similar in appearance are the ‘ horned
toads’ (Phrynosoma) of the desert regions in the
warmer parts of the United States and in Central
America. They also have a rough and prickly skin,
and possess remarkable powers of burrowing. The
animals are assisted in the process by a curious fringe
at the sides of the body, and are said to appear actually
to sink into the sand. They feed upon a variety of
insects, and in colouring resemble the sandy soil in
which they live. In the hot deserts of Northern Africa
and the adjacent regions occur the skinks (Scincus),
which are admirably adapted for desert life, their
bodies enabling them to move through the soft sand
as fish swim through water.

The presence of the lizards named, together with
many others, speaks to the abundance of insects in
desert and savana regions. These insects do not, as
a rule, show any very marked adaptations, but we may
mention the termites, or so-called white ants, as being

AND DESERTS 143

very characteristic of the savana regions of Africa and
elsewhere.

Rerzrences. All the books mentioned in the previous chapter as
giving accounts of African animals, describe the inhabitants of the
savanas as well as those of the forests. The books mentioned at the
end of Chapter X also, as giving accounts of the zoological regions,
describe among others the savana animals. Mention may also be made
of W. H. Hudson’s books, especially The Naturalist in La Plata (London,
1892); and of Tristram’s Flora and Fauna of Palestine (London).

CHAPTER VII

SPECIAL FEATURES OF ISLAND. FAUNAS

IsLANDS present some interesting features as regards
their animal life which makes them worthy of special
study. We have already seen that mountain areas have
special types of animals, because they form isolated
regions cut off by the physical conditions from the
neighbouring districts. The special conditions in their
case include both climate and topography, and in what-
ever region of the globe the mountains occur, if they are
greatly elevated, the climatic conditions are similar.
Islands are regions cut off from neighbouring regions
by the sea, which forms a barrier to the passage of all
but the marine and the flying animals. Even to flying
animals a large expanse of sea may constitute a for-
midable barrier, save in the case of those with excep-
tional powers of flight. The consequence is that island
faunas form a whole, and can be studied with much
more ease than the animals of any other natural region.
The question whether a particular animal does or does
not occur in an island is merely a question of observa-
tion; while that as to whether a form found on a
mountain does or does not constitute part of the moun-
tain fauna is a much more delicate one, involving an
elaborate process of reasoning. In the one case a
gradual migration with concurrent adaptation is pos-
sible from plain to mountain or vice versa, but no such
slow migration is possible in the case of islands.

On the other hand, it is to be noted that while

ISLAND FAUNAS 145

mountain animals, like those of steppes or forests, are
exposed to virtually uniform conditions of climate, &c.,
no such uniformity exists in the case of islands. We
speak, it is true, of insular climates, meaning that the
proximity of the sea has a certain moderating effect,
more especially if there is a large expanse of ocean all
round, or if the prevailing winds blow from such an
expanse. But even within the same island there may be
great variations of climate, and there may be almost all
types of surface—mountain, wood, steppe, tundra, and
so forth. Thus, while the inhabitants of islands may
be readily counted and classified, and form a statistical
unity, they do not form a biological unity in the sense
that the animals of the other natural regions do, and
there are few general statements which can be made
of them as a whole.

The special features of island faunas may best be
illustrated by a few examples, and we may begin by
a consideration of the Galapagos Islands, an interesting
group which has been studied by various naturalists,
including Darwin.

This archipelago consists of a group of islands lying
some seven hundred miles off the coast of Ecuador,
and on the equator. They are of volcanic origin, the
surface being covered with extinct craters, and are
placed upon a submarine bank, which rises steeply
from the depths of the Pacific. The climate is singu-
larly cool for the position, this being due to the fact
that the islands are washed by the cold Humboldt
current. In the lower regions of the islands the rain-
fall is markedly deficient, rain only falling in small
amounts in the earlier months of the year (February—
May). Here the vegetation is scanty, and xerophytic
in character, the most important plant from the

1404 K

146 SPECIAL FEATURES

economic point of view being the archil lichen from
which the dye called litmus is obtained. Darwin
describes this lower zone as uninviting in the extreme,
the scanty brushwood leaving the black basaltic lava
exposed, and giving the district the appearance of an
iron-smelting region in the ‘ black country ’ of England.
But the clouds hang low, and above an elevation of
about eight hundred feet not only is there abundant
rain during the rainy season, but even at other seasons
a constant mist keeps the vegetation fresh. The plants
of the two regions differ fundamentally, and in the
upper cultivated plants from many climates thrive. Of
the native vascular plants, about one-half are peculiar
to the islands. The group consists of five principal
and eight small islands, with a total area of nearly
3,000 square miles. The hills range in height from
3,000 to 4,000 feet. The more conspicuous bushes and
trees are species of euphorbia, cactus, and acacia,
so that the islands generally resemble savanas and
desert regions in their vegetation. Palms are entirely
absent.

Turning now to the fauna, we find that mammals are
practically absent. There are two mice, but of these
one belongs to the widely distributed genus Mus, and
has doubtless been introduced ; the other belongs to
a South American genus (Hesperomys), and has also
possibly been introduced. That there is nothing in the
conditions of the islands to prevent the existence of
even large herbivorous mammals is, however, shown by
the fact that goats, pigs, horses, &c., introduced by
man, have become wild, and have multiplied exceed-
ingly.

The birds present some interesting features. About
seventy species, according to Wallace, have been

OF ISLAND FAUNAS 147

obtained in the islands, and of these forty-one, or con-
siderably more than half, are peculiar to the islands,
not being found elsewhere. Most of those which are
not peculiar are birds with considerable powers of
flight, capable of travelling great distances: Of the
true land birds all but one are peculiar, and this one
is the common rice-bird (Dolichonyx oryzivorus), which
occurs throughout the whole of the American continent.
This bird is markedly migratory, spending its winters
in the West Indies and Central America, and travelling
northwards in vast numbers to breed and spend the
summer in the northern United States and Canada.
There is, therefore, nothing improbable in the supposi-
tion that stragglers from the migrating flocks reach
the Galapagos Islands from time to time, and so pre-
vent an island type from establishing itself. The other
land birds of the islands show, as Wallace points out,
all gradations from close resemblance to forms occurring
elsewhere to generic difference. Thus the archipelago
contains a special owl—Asio galapagensis—but this is
very nearly related to the widely distributed short-
eared owl (Asio brachyotus). On the other hand, among
the finches peculiar genera occur, e.g. Geospiza, related
to genera of restricted range found on the continent
of South America. Again, though the islands are
relatively so near together, a peculiar genus may be
represented by three different species on as many
different islands. This is well exemplified in the case
of the honey-creepers of the genus Certhidea, the three
species of which occur on different islands.

The reptiles of the islands are especially remarkable.
The giant tortoises have long been famous, and give
the islands their Spanish name. Such giant tortoises
occur only on islands, being found in certain islands in

K2

148 SPECIAL FEATURES

the western Indian Ocean, and in the Galapagos archi-
pelago. In addition to their large size the animals have
as special peculiarities the fact that they show much
individual variation, and tend also to split up into
species, different species occurring on adjacent islands.
No small tortoises occur on the islands containing these
large forms. A final peculiarity is that the bony cara-
pace tends to be reduced in thickness, perhaps because
till man came the animals had few enemies. The diet
is exclusively vegetarian, and the animals occur in both
the upper well-watered region and in the lower arid
region. The fact, however, that they drink greedily,
and are fond of wallowing in muddy water, suggests
that they are not native to arid regions. Darwin gives
an interesting description of the long journeys made by
the low-ground tortoises to the springs of the upper
region, in order to obtain water. The cactuses of the
arid regions form a favourite food plant. The exact
number of species is a matter of doubt, owing to the
individual variability already mentioned.

In addition to the giant tortoises, the Galapagos
possess both lizards and snakes. Two of the lizards
are very large and of very peculiar habits. One (Cono-
lophus subcristatus) is found only on the central islands
of the archipelago, and occurs both in the upper and
lower regions, but more abundantly in the latter. The
forms from the arid districts, however, do not travel
upwards to the springs to drink, and are far more
numerous than the upland forms. As the diet of these
lizards is quite similar to that of the tortoises, it would
seem that originally the lizards inhabited the arid
ground, and the tortoises the damp upper regions, but
each has become partially adapted to the territory of
the other. The lizards reach a length of a yard, and

OF ISLAND FAUNAS 149

have stout heavy bodies. They are very partial to
cactuses, but also climb the acacia trees to feed upon
the foliage.

Closely related to Conolophus, and therefore like it
one of the iguana family, is the marine lizard Ambly-
rhynchus cristatus. It also is entirely peculiar to the
islands, and is the only lizard known which is marine
in habitat. It inhabits the shores of most of the islands,
and feeds upon seaweeds which grow below tide mark,
and, as Darwin proved, it can tolerate prolonged im-
mersion without any ill effects. It will be recollected
(cf. p. 116) that the typical iguanas are forest animals.
The exceptional conditions to which the forms in the
Galapagos Islands are exposed has made it necessary
for these to adopt modified habits.

The family Iguanidae is further represented in the
islands by various peculiar species of the genus Tropi-
durus, other members of which occur on the continent
of South America. Various species of geckoes belong-
ing to the widely distributed genus Phyllodactylus also
occur, most of these being peculiar to the islands.
There are also two snakes, both being nearly allied to
South American forms.

Amphibians are entirely absent. Of the inverte-
brates the insects are few in number, but show the same
characters as the other kinds of animals. In other
words, most of the species and many genera are peculiar,
and the forms tend to have a limited distribution
among the islands, very few being found in all the
islands. Further, there are many wingless forms,
a common feature in island insects.

It is not necessary to describe the fauna of other
islands of the same type as the Galapagos archipelago
in detail, for the general characters are the same in

150 SPECIAL FEATURES

all. Islands which do not lie very near a continent,
and are separated from the nearest continent by
deep water, that is, do not lie upon the same con-
tinental shelf as the continent, show great general
similarity as regards their fauna. In such islands there
are no indigenous land mammals, often with the doubt-
ful exception of mice and rats, animals readily intro-
duced by man. Amphibians are also absent, a fact
readily explained when we recollect that both the eggs
and adults are very intolerant of salt, and therefore
could not be readily transported across sea-water by
floating logs of wood or any similar means. In such
islands generally the commonest land forms are birds
and invertebrates, especially insects and land shells.
These in all cases tend to run into local races, due
apparently to the isolation of the stock. Further, fly-
ing forms tend to lose their wings or to possess only
limited powers of flight. This is very marked in some
birds, thus the flightless dodo, the solitaire, the kiwi of
New Zealand, and so forth are all inhabitants of islands.
It is noticeable that the power of flight tends specially
to diminish in animals found in islands much exposed .
to hurricanes, for here even short flights might expose
the animals to strong winds, which would sweep them
out to sea, and thus destroy them, while those which
flew little or not at all would be more likely to breed
and continue their own type.

Reptiles are not very frequent in islands separated
by deep water from continents, and the presence of
the large tortoises in the Galapagos Islands is held by
some to indicate a connexion, at a remote period, with
the continent of America. Wallace, on the other hand,
believes that the ancestors of these forms were acci-
dentally introduced, possibly on floating timber.

(‘unasnyy yseyoog pohoyy
ayy us uauaads » wotg) ‘AzyuNoo yeyy OF reynoed priq ssopyysip v ‘pueeog Many Jo (whuajdy) tury on, “OF “PML

OF ISLAND FAUNAS 151

Generally the fauna of islands like the Galapagos, called
oceanic by Wallace, is such that its members may
be supposed to have been, at least to a large extent,
accidentally introduced—by storms carrying winged
animals out of their usual course, or by mechanical
transportation on floating wood, &c. Incidentally we
may note that the absence of mammals on oceanic
islands has often important human consequences, for
the paucity of sources of animal food tends to make
their human inhabitants cannibals.

As contrasted with oceanic islands we have the
islands which are obviously merely separated portions
of the adjacent continent upon whose continental shelf
they stand. A good example is furnished by the British
Isles, whose fauna is very similar to that of the adjacent
continent, but shows impoverishment in several re-
spects, apparently as a result of the glacial period.
As Ireland was apparently cut off from the continent
before Great Britain, and at a stage when many of the
animals driven south by the ice had not had time to
recover lost ground, we find that its fauna is poorer
even than that of the larger island. The following
figures, quoted from Wallace, will help to render the
above statements more precise. Scandinavia possesses
about 60 species of mammals, Great Britain 40, and
Ireland 22; Belgium has 22 species of reptiles and
amphibia, Great Britain 13, and Ireland only 4. The
question as to the number of peculiar forms in the
British fauna has been actively debated. That the
red grouse (Lagopus scoticus) of the northern regions is
peculiar is generally admitted, though it is nearly allied
to the willow grouse of Scandinavia. We have also
some peculiar fresh-water fish, most of which occur in
lakes. The differences are, however, not very striking,

152 SPECIAL FEATURES

and are no doubt due to the fact that the chances of
the fish of one loch interbreeding with those of another
are very small. There are also a considerable number
of peculiar species or varieties of insects, but nothing
comparable to the differences which separate the
animals of the Galapagos Islands from those of the
adjacent mainland. To add point to the contrast we
may note that the area of the British Isles is in round
numbers, 121,500 square miles, as contrasted with the
3,000 square miles of the Galapagos Islands. Suchislands
as the British Islands are called by Wallace continental,
in contrast with the oceanic type.

One more example of an island fauna may be given.
In this case we shall take Madagascar, which is of the
continental type, but is separated from the continent
of Africa by a channel so deep and wide that it must
be supposed that the connexion between the two was
broken at an extremely remote period. The total area
of this large island is nearly twice that of the British
Isles, and there is much forest. The fauna is remark-
ably rich, and is as markedly characterized by the
animals it includes as by those of which it is devoid.
Thus the island has no monkeys in its wide tropical
forests, but has some thirty-three species of lemurs,
which constitute half the mammalian population
of the island. Insectivores are fairly numerous, and
include in the tailless hedgehogs (Centetes) primitive
forms whose nearest allies appear to be some shrews
in the West Indian Islands. There are not many
carnivores, and no true cats, but in addition to eight
civets there is a relatively powerful animal called
Cryptoprocta, about the size of a common cat, which
belongs to the civet group. There are few rodents, and
the enormous wealth of large ungulates, which is so

OF ISLAND FAUNAS 153

characteristic of the continent of Africa, is here scarcely
suggested, for that important order is only represented
by a river-hog and a subfossil pygmy hippopotamus.
Both the species and genera of mammals are strikingly
peculiar; thus it appears that of the twenty-four
mammalian genera of Madagascar only two occur in
Africa.

The birds are not quite so peculiar, but they also
show, though to a less degree, the same feature of the
absence of the African forms which would be expected,
and the presence of primitive or peculiar forms. Among
both snakes and lizards we find the peculiar feature
that American forms are represented ; among the lizards
by two Iguanidae, otherwise peculiar to America. The
other lizards are mostly peculiar members of African
families, but a special feature is the number of
chameleons, and the absence of the monitors, which
are abundant in Africa. There are no poisonous
snakes. The amphibians are remarkable, for they
present affinities both with the Indian region and with
South America. The beetles, again, show affinities with
those of India, Australia, and the Malayan region.

Without considering in detail the various questions
raised by these peculiarities, it may be sufficient to say
generally that Madagascar was formerly connected to
Africa, and perhaps also through a chain of islands
to India. The connexion was broken before the higher
forms of life, e.g. the monkeys, the higher ungulates,
&c., had penetrated Africa, and consequently the
original mammalian and other fauna of Madagascar
was primitive in type. Saved from the competition of
higher forms by isolation, and favoured by the abundant
food, the absence of enemies, and the great forest area,
many forms, notably the lemurs, have reached a high

154 SPECIAL FEATURES

degree of specialization along their own lines, and have
evolved in many different directions.

If we sum up the more important general points in
regard to island faunas, we may note first that Wallace’s
distinction into oceanic and continental islands is not
of very great practical importance, because the pro-
gress of geology, since the classification was laid down,
has made it difficult to apply. It involves a conception
of the permanence of ocean basins which is not now
generally accepted. Thus, while Wallace names the
Galapagos group as a characteristic oceanic group,
others believe that the presence of giant land tortoises
is in itself a proof of a former land connexion. For
our purpose then it is sufficient to recognize the follow-
ing three conditions :

1. When an island has never had a connexion with
a continent, e.g. many coral and volcanic islands, or
has only had such a connexion at a geologically remote
period (Galapagos ?) before the higher forms living at
the present day had originated, or before they had had
time to become widely distributed, its fauna contains no
mammals, no amphibia, (1) no reptiles, or (2) reptiles of
avery primitive type, or (3) such as could be accident-
ally carried by floating wood, &c. The most important
members of its fauna will be birds, insects, and land
shells. Owing to the isolation the animals will tend to
exhibit special peculiarities, fitting them for their
habitat. Among flying forms the conditions favour
those in which the power of flight is moderately de-
veloped, or has disappeared. Examples of such islands
are the Azores, the Bermudas, St. Helena, the Sand-
wich Islands, &c. Where such islands have had a
continental connexion, this must have been prior to
the deposition of the lower Tertiary beds, for at this

OF ISLAND FAUNAS 155

time mammals were evolving rapidly and spreading
widely.

2. When an island has been separated from adjacent
land masses for a moderately long time, i.e. since the
middle of the Tertiary period, it will tend to have
a rich fauna, including representatives of all the chief
classes and phyla, but the more modern classes will be
represented by primitive forms, their primitiveness
depending on the length of the period during which the
island has been isolated. Of such islands Madagascar
and the adjacent islands of the Indian Ocean are good
examples.

3. When the separation of an island from the adjacent
continent has taken place within geologically recent
times, i. e. in the Pleistocene or Post-Pleistocene period,
the fauna will be in all essentials similar to that
of the adjacent land mass, though minor differences
may occur. Examples are the British Isles, Japan,
Borneo, Java, &c.

In other words, the most satisfactory classification
of islands from the point of view of their fauna depends
upon the approximate length of time since which they
have been isolated.

REFERENCES. Wallace’s Island Life (second edition, London, 1892) is
the classical work dealing with the subject, which is also discussed to
a minor extent in the same author’s Darwinism. A very interesting
account of the Galapagos Islands will be found in Darwin’s Journal of
Researches into the Natural History and Geology of the Countries visited
during the Voyage of the ‘ Beagle’ (London, 1843). The subject is also
fully treated in Beddard’s Tezt-book of Zoogeography (Cambridge, 1895).
See also Dobson’s, On Some Peculiarities in the Geographical Distribution
and in the Habits of certain Mammals inhabiting Continental and Oceanic
Islands (Ann. Nat. Hist. xiv. 1884).

CHAPTER VIII

THE DISTRIBUTION OF ANIMAL LIFE
IN THE SEA

WHILE, as we have seen in the previous chapters,
the natural regions of the land more or less grade
into one another—savana, for example, passing into
savana wood, and savana wood into forest, it is very
much easier to separate the oceans into natural regions,
for here the divisions are relatively sharp. The two
great conditions which influence the type of animal to
be found within any part of the sea are, first, the pres-
ence or absence of light, and, second, the presence or
absence of asubstratum. At the margin of the sea, where
the light penetrates to the bottom, we have the littoral
area, peopled by littoral organisms. All these organisms
agree in that, in structure or in habit, they show adapta-
tions to the two striking features of the area. Thus
a sea anemone, fixed to a rock and containing symbiotic
algae, is obviously adapted for life in a region where
a substratum is present, and where light penetrates.
A plaice, so shaped that it is adapted for lying on
a sandy surface and bearing well-developed eyes, is
similarly fitted. But the dependence of a mud-haunt-
ing animal feeding upon minute algae upon littoral -
conditions is no less real, if less apparent, for land-
derived mud only accumulates off shores, and the
minute algae must have light before they can live.

In the open water occurs another group of organisms,

ANIMAL LIFE IN THE SEA 157

constituting the pelagic forms, which require no sub-
stratum, but for the most part demand light, and usually
display sensitiveness to it. Finally, at the bottom of
the ocean, in the great depths, we have forms entirely
independent of sunlight, but directly or indirectly
dependent upon the presence of a substratum. These
are the abyssal animals, whose existence was first fully
demonstrated by the Challenger expedition.

We can thus divide the inhabitants of the ocean into
three well-defined classes, each class showing adapta-
tions to a natural region. These three groups are
(1) the littoral, (2) the pelagic, and (3) the abyssal,
each of which we must consider separately. As we
shall see, the members of the different groups do not
necessarily restrict themselves throughout their whole
life history to one of the natural regions. It is, for
example, very common to find that littoral forms have
pelagic larvae, or pelagic stages in their life-history,
able to take advantage, for purposes of distribution,
of the many currents which influence the surface waters
of the sea.

(1) The littoral zone extends outwards from the
margin of the land to the edge of the continental shelf,
or roughly to the one hundred fathom line. Beyond
this line the sea-bottom usually slopes rapidly in the
Continental Slope to the great depths, where the
littoral animals are replaced by the abyssal ones.
Shallow seas, such as the North Sea, contain only
littoral and pelagic animals. Throughout its exten-
sion the littoral area is characterized by its great
wealth of food, and its great variations in the physical
conditions. The basal sources of food are here three :
(1) the waste of the land, (2) the fixed algae of the
shore rocks, (3) the minute floating algae or phyto-

158 THE DISTRIBUTION OF

plankton. Theimportance of the waste of the land may
be realized if we think of the number of marine animals
which find food in the mud-deposits which accumulate
round the mouths of rivers and estuaries. If the water
be fairly clear, such estuaries are specially favourable
to shell-fish like mussels and oysters ; to many marine
worms, which in their turn feed many fishes ; to many
small crustacea, which also feed fish ; and so forth.
This land-waste has only a limited seaward extension,
and is not available as a source of supply either for the
pelagic or for the abyssal animals.

The fixed algae can only grow where some firm sur-
face to which they can attach themselves exists, and
where the sunlight is strongly felt. They are therefore
limited to the margin of the lands, but there form rich
feeding-grounds. The Laminarian zone, which lies just
beyond low-tide mark, has always been known as
a rich collecting-ground. It contains many small forms
—crustacea, molluscs, worms, coelenterates, and so
forth—perfectly adapted to life among the long brown
fronds and about the branching roots. At high tide
this zone is covered with a sufficient depth of water
to be available as a pasture ground even for the larger
fish, and here therefore shore animals cluster, and the
fisherman reaps rich harvests.

Finally, in the shore waters minute algae swarm.
In their habits and adaptations most of these are
pelagic, that is apparently independent of the presence
of a substratum, but the fact that there is considerable
difference between the phytoplankton of the shore
water and of the open sea, suggests that these forms
are in some way indirectly affected by the proximity
of the sea-floor, or by the waste of the land. Diatoms,
especially, which are eagerly consumed by many

ANIMAL LIFE IN THE SEA 159

animals, are represented in the shore waters by many
species—by more than in the open ocean. But, it
may be asked, if these algae have no attachment to
a substratum, how do they contrive to remain in
the shore waters, in spite of currents and tides which
must tend to sweep them seawards? The answer
is apparently that myriads are so swept out into
the ocean, but there they degenerate, and probably
do not live long. Produced in the shore waters, their
destiny may be to serve as food for the pelagic animals
of the open seas, but they no more form a true part
of the life of the pelagic area than does that shore-
born weed which floats at the surface of the Sargasso
Sea for a period, ere it perishes and is replaced by new
fragments torn off by the currents.

Of the physical peculiarities of the littoral area the
constant movement of the waters is characteristic.
Tides and currents, in all the open seas, ensure that no
stagnation shall occur, and thus keep the water sweet,
and bring constant supplies of food and oxygen. It is
this constant movement which helps to give the shores
of the oceans so much richer a fauna than the shores of
lakes, or of enclosed seas like the Mediterranean or
Black Sea. The constant movement brings with it
also a danger, whose influence is manifest in several
ways. Some of the shore forms are swift swimmers,
strong to stem the tide, and yet, notwithstanding their
strength, the shore naturalist knows that off many
coasts he is sure of a rich harvest when wind enforces
tide or current, for the beach at certain seasons is
strewn with flotsam and jetsam. Many shore fish, the
strong cuttles among molluscs, powerful forms like
porpoises and dolphins and even whales—there are
times when the strength of these avails them nothing

160 THE DISTRIBUTION OF

against the movement of the waters, and they are found
dead or stranded on the beach.

The weaker abandon the attempt to struggle against
the waters—the lumpsucker and the gobies among fish,
the limpet among molluscs, the acorn shells and bar-
nacles among crustacea—these do not attempt to resist
the tides, but cling tight and allow the danger to pass.

Another danger, quite as real, though not so obvious,
which the shore forms must guard against is the con-
stant change of temperature and of salinity in the sea
water, which is characteristic of the shore. The
temperature of the oceans varies much, and is in-
fluenced by local causes, but the shore waters especi-
ally are greatly affected by the variations in the tempera-
ture of the land, that is by latitude. In tropical waters
the range of temperature is small (about 10° F.), and
the mean temperature is high (about 80° F.). In polar
waters, while the range is similarly small, the mean
temperature is low (about 28° F.). Between the two,
in temperate regions, the range is considerable, while
the mean is. neither high nor low. Nowhere, of course,
can the ranges of temperature in the sea be compared
with those on land, owing to the high specific heat of
water. We find, however, that very many land animals,
for instance all the warm-blooded ones, have a heat-
regulating mechanism, whereby they can adjust them-
selves to changes of temperature. Such a mechanism
is absent in all sea animals except those (e. g. whales,
&c.), who are descended from terrestrial ancestors.
According as they can or cannot tolerate variations
in the temperature of sea water, sea animals may
be divided into eurythermal and stenothermal forms.
The former are tolerant of great variations, the latter
are rapidly killed by them. Speaking generally, the

ANIMAL LIFE IN THE SEA 161

littoral animals of temperate latitudes must be eury-
thermal, while those of tropical and of polar seas may
be stenothermal. As the actual temperature in the
case of many stenothermal animals seems to matter less
than its constancy, the temperate zone must form
a barrier between the animals of the two other zones,
and more or less prevent their mixing. Further, as
sudden changes are always more dangerous than gradual
ones, the contact of a cold current with a warm one
seems often to lead to a great destruction of animal
life, though this perhaps affects pelagic animals more
than littoral ones.

Variations in the salinity of the water seem to have
similar effects, and Hjort states that of the seventeen
species of the cod family found in the North and
Norwegian Seas, each one seeks out a special spawning-
ground, having its characteristic depth, temperature,
and salinity, a fact which may be partly due to greater
susceptibility of the larvae than of the adults to minor
variations in pressure, temperature, and salinity.

Something must now be said as to the characters of
shore animals. In the chapter on the tundra we men-
tioned some of the common forms to be found in Arctic
regions, and to that account some more general state-
ments may be added. Of the marine mammals we
must regard all the seals and the walrus as littoral
forms, for though many show a perfect mastery of the
water, all come on shore for breeding purposes, and
thus are dependent on a substratum, at least at certain
seasons. In addition to the seals already mentioned,
we have the true fur-seals (Otaria), which occur in the
north, outside the Arctic area, and the many seals of
the Far South, such as Weddell’s seal (Leptonychotes
weddelli), Ross’s seal (Ommatophdca rossi), the crab-

1404 L

162 THE DISTRIBUTION OF

eating seal (Lobodon carcinophaga), the sea-leopard
(Stenorhynchus leptonyx), which all occur in the region
of ice-pack, while the sea-elephant (Macrorhinus leo-
ninus), and the southern fur-seal (Otaria jubata) occur
in sub-Antarctic seas as well as elsewhere. The fact
that all these different kinds of seals tend to occur in
cold regions cannot be ascribed wholly to deliberate
choice on their part, for the young of all are born on
the land, and the animals seek out places as free as
possible from molestation, which are more frequent in
polar seas than elsewhere. Since man joined their
enemies many species are being driven further and
further back into the inaccessible parts of the polar
seas, or have been virtually exterminated.

The seals, walrus, and sea-otter, the last named
being a fur-bearing animal found off rocky coasts in
the North Pacific, are all carnivores which have taken
recently to the aquatic life, and are still bound to
terra firma at the breeding season. There are, how-
ever, twaq orders of mammals whose members are
entirely aquatic, their young being born under water.
These are the Cetacea or whales and dolphins, and
the Sirenia or sea-cows. The former are powerful
swimmers, and are pelagic. Their only connexion with
shore is that the greater abundance of fish in shallow
water brings many of them to the vicinity of sub-
marine ‘banks’, while the whalebone whales haunt
relatively shallow water in high latitudes, because of
the abundance of small marine organisms there.

The Sirenia include only two living animals, the
dugong (Halicore) found in the Pacific and Indian
Oceans, and extending southwards to the shores of
Australia, and the manatee (Manatus) which lives in
the estuaries of the great rivers which flow into the

' Fig. 41. The shore of Copper Island, in the Bering Sea, showing a Rookery of Fur-seals.

The seals prefer the rocky portion of the beach close to the water, and only the weaker
‘forms, unable to obtain a foothold there, haunt the smooth ground at the back of the
rookery. (Photo by Prof. D’Arcy W. Thompson.)

(-uosdwoy,, -M Soap. ‘ford fq royd) “09s
et} Ul Os[e pure clogs A[suTYs ey} UO BLEEy Surwoys ‘pussy seddog yo youoq oy} Jo WUT “BP OL

ANIMAL LIFE IN THE SEA 163

tropical part of the Atlantic, and in shallow bays or
lagoons. Though both spend their whole life in the
water, and are helpless outside it, yet both are definitely
littoral, for they feed upon the large algae, and could
therefore find no food in the open sea. The manatee
apparently takes a large variety of aquatic plants in
addition to algae. The animals are not nearly such good
swimmers as the Cetacea, and their heavy bodies and
massive bones fit them for life near the bottom. The
hind-legs are entirely absent, but the fore-limbs may
touch the ground as the animals swim leisurely along
the bottom, and are also used in pushing food into the
mouth, and in carrying the young, which are suckled
above water.

Of the littoral birds perhaps the most interesting are
the flightless penguins, which present the same features
as the seals in that, though pelagic for much of the year,
and having a perfect mastery of the water, they must
come on shore for breeding purposes. The absence of
any carnivorous land mammals in the Antarctic area,
and its inaccessibility, make it a perfect paradise for the
penguins, who resort to its rocks and beaches in millions
to breed, and have been described by all the Antarctic
expeditions (see Fig. 44). Quite comparable in habits,
though not related, was the extinct great auk (Alca
impennis) of the shores of the North Atlantic, which
was similarly incapable of flight, and fell an easy
victim to man during the breeding-season.

Of the other birds which feed in the sea and frequent
it, some, like certain of the seagulls, forsake it at the
breeding-season for inland regions. Others, like little
auks, guillemots, puffins, gannets, and so forth, breed
on the shore, choosing cold or inaccessible regions
for the purpose. Of the various adaptations to such

L2

164 THE DISTRIBUTION OF

breeding-places which they show the commonest is the
pointed egg, which does not roll off the rocks readily,
while the eggs themselves are frequently protectively

Fic, 43. Kittiwake Gulls nesting on basaltic pinnacles on May Island,
in the Firth of Forth. (Photo by Mr. George West.)

coloured, except where, like those of the puffin, they
are concealed in holes.

Of the reptiles relatively few are marine animals,
the ‘habit being most frequent among the turtles and

(-sonag “g" M “4 Aq oy) “Summon sumBueg “FF “PML

ANIMAL LIFE IN THE SEA 165

tortoises (Chelonia). The very curious leathery turtle
(Sphargis coriacea) is apparently found in all tropical
seas, but it is either very rare or its habits make it
infrequently seen and caught. It is a large animal,
sometimes measuring six and a half feet long, and feeds
exclusively on fish, molluscs, and crustacea. The limbs
are turned into paddles, the anterior ones being very
long, and the animal is apparently pelagic in habit and a
swift swimmer, though it is tied to the land by the need
of breeding there. The eggs are laid on sandy shores,
especially on the eastern side of tropical America.

The green or edible turtle of commerce (Chelone
mydas) is also marine, but the fact that it is vegetarian,
feeding upon algae and other aquatic plants, binds it
closely to the shore, though the animals are good
swimmers. As usual the eggs are laid on sandy beaches.
Another species of the same genus is the hawksbill
turtle (Ch. imbricata), which supplies the tortoiseshell
of commerce, and is carnivorous, feeding upon fish and
molluscs. Both these forms have a wide distribution
in warm seas, while the loggerhead (Thalassochelys
caretta) extends into the Mediterranean and Bay of
Biscay. This is a carnivorous form, and occurs hun-
dreds of miles from land, floating on the surface. Again,
however, it must approach the land at the breeding-
season.

We have already spoken of the marine lizard of: the
Galapagos Islands (p. 149), and there are also sea-
snakes (Hydrophis) similarly adapted for life in the
sea. These animals, which are especially found off the
coasts of the Indian Ocean, have compressed swimming-
tails, and feed upon fish. As in the case of so many
forms which have secondarily returned to the aquatic
life, their dependence on the land is especially shown

166 THE DISTRIBUTION OF

at the breeding-season, for the gravid female must come
on shore to produce her young. For this purpose she
visits the shores of low islands, where she may be found
coiled up among the rocks.

The fact that amphibians have unprotected skins,
through which salt can pass very easily, makes them
intolerant of its presence, and we have thus no marine
amphibia.

Littoral fishes are many, and range from forms like
the lumpsucker, the gobies, the sea-horses, the pipefish,
and so forth, which inhabit rocky pools within or close
to tide marks, through forms like plaice, dabs, turbot,
skate, fishing frogs, &c., which haunt the sandy bottom,
their shape fitting them for life here, to forms like cod
and haddock, whose chief adaptation to littoral life is
_ that they are ground feeders, depending upon animals
like hermit crabs, marine worms, shellfish, &c., which
only occur on the bottom.

It is a curious illustration of what has been already
said as to the tendency to change from the pelagic to
the littoral habitat during the life-history, that the
littoral cod and haddock should have floating or
pelagic eggs, and should in their early life haunt the
surface, often seeking shelter within the bells of the
pelagic jellyfish, while the herring, which in adult
life is a pelagic fish, feeding on free-swimming crus-
tacea, has heavy or demersal eggs, which sink to the
bottom.

The tunicates or sea-squirts mostly attach them-
selves to the sea-bottom, but have pelagic free-swim-
ming larvae.

Of the molluscs a vast number are littoral forms.
Many cuttles cower down on the sea-bottom waiting
for their prey, or clamber over the rocks by the aid of

ANIMAL LIFE IN THE SEA 167

their sucker-bearing arms. The flat creeping surface
which forms the foot in most gastropods, such as peri-
winkles, whelks, and so forth, indicates the dependence
of these animals upon a substratum. The bivalves
often live buried in the mud, like Mya and Lutraria,
or in sand like cockles, or they are attached by a byssus
or tuft of silky threads like the common mussel, or they
bury themselves in the substance of the rocks like
Pholas and Saxicava. Many of these forms, however,
have pelagic larvae.

While many of the crustacea are pelagic, like the
copepods or water-fleas, and various forms of prawns,
others, like crabs, are adapted for life on the bottom.
Indeed, we can trace among the shore crustacea the
gradual acquirement of the ground-haunting habit,
leading by various stages from a form with the power
of swift swimming like the lobster, to forms which can
only walk like the crabs, these showing many very
interesting adaptations to their relatively sedentary
life.

The majority of the annelid worms are littoral,
relatively few being adapted to the pelagic life. Curiously
enough, however, not only have most of them pelagic
larvae, but certain forms become temporarily pelagic at
the breeding-season. The most striking example of this
is the palalo worm of various islands in the Pacific.
This worm lives among the coral reefs, but at certain
seasons a portion of the body containing the sexual
elements is liberated, and leads a brief free-living life
at the surface, before death occurs. The liberation
of these sexual portions takes place at night, simul-
taneously in thousands of worms, so that the whole
surface of the water becomes filled with the animals,
which off Fiji, the Friendly Islands, and elsewhere,

168 THE DISTRIBUTION OF

are eaten by the natives as food. The portion of the
worm which remains behind in the coral reefs regener-
ates the missing segments, and a year later the process
is repeated. A similar process is suggested in certain
British annelids, and is a device to ensure that the
eggs are carried to a distance from the parent.

The littoral annelids show many beautiful adapta-
tions to shore life. We can only name the sea-mouse
(Aphrodite), which lives buried in mud; Serpula,
which secretes lime from the sea water and fashions
a tube; which it attaches to rocks or to the hard parts
of other animals or to weed; Terebella, which builds
up tubes from the sand, and implants them deeply in
the sand so that it is safe from the wash of the waves ;
a Nereis, which lives inside the shell inhabited by
a hermit crab, and thus finds protection from its
enemies, and so forth. Very many of these annelids
are greatly prized as food by fishes, and it is their
abundance in the shore waters that attracts so many
fish there.

Very many of the echinoderms also are littoral, and
manifest obvious adaptations to this mode of life. The
sea-urchins burrow in the sand or mud, or else climk
about the rocks by means of their suctorial tube feet.
Starfish and brittle-stars similarly clamber over the
bottom, while the holothurians either live buried in
the sand or attach themselves by their tube feet.

Not a few unsegmented worms are littoral, a pretty
example being the turbellarian called Convoluta, which
possesses symbiotic algae, a common feature of shore
animals of simple structure.

Perhaps the most interesting forms are, however, the
Coelentera, which often show a double adaptation to
littoral and to pelagic life. The fronds of the shore

ANIMAL LIFE IN THE SEA 169

algae are often covered with the delicate tracery of
the sea-firs, colonial coelenterates of sedentary habit.
Among the branches of the ‘ fir’ special cups may be
seen in which buds form. These buds drop out and
float away, becoming swimming-bells—sexual forms
which carry eggs and sperms to a distance, so that the
fertilized egg may find a home far from the parent
' stock. This, which the biologist calls alternation of
generations, is of common occurrence among the
coelenterates. We find, however, that frequently one
of the two modes of life is emphasized at the expense
of the other. Those delicate bells which we call Cteno-
phora, or comb-bearers, have no littoral stage at all,
but are pelagic throughout their life. On the other
hand, sea-anemones, sea-pens, and their allies, the
corals of warm seas, and so forth, are littoral, save in
their earliest stages, and are fixed save in those very
early stages. No swimming-bell floats away from
the coral as it does from the sea-fir colony. Between the
two extremes all stages exist. For example, many of the
great jellyfish are shore-born, spending a considerable
part of their lives attached to rocks. Others are pelagic
throughout, while still others remain permanently
attached to the rocks, like larvae which have never
grown up. Somewhat similar conditions occur in the
tunicates already mentioned (p. 166).

The sponges, which all require the presence of a sub-
stratum, are necessarily either littoral or abyssal.
A great number occur in shallow water, all these being
attached to some solid body. But there are no very
obvious adaptations to littoral life, save perhaps that
shallow-water sponges seem to be always attached,
while some abyssal forms lie loosely on the surface of
the ooze.

170 THE DISTRIBUTION OF

The protozoa are as well represented on the shore as
in other parts of the sea, but present no special adapta-
tions to life here.

(2) We come next to the pelagic region, where there
is no substratum, but where many at least of the
organisms are fully exposed to the action of light.
With regard to the movements of the water here, we
must notice first that, in contrast to the littoral area,
tides are of little importance. On the other hand,
currents are very important. As is well known,
the difference in the amount of heat which streams
down from the sun upon the surface of the sea gives
rise to differences in temperature and density, which
cause the surface waters to move in great whirls, the
ocean currents. As the surface in the different latitudes
is exposed to different types of winds, and therefore to
differential evaporation, we have also differences of
salinity. Roughly speaking, the salinity is greatest in
warm regions exposed to the drying trades, e. g. in the
Red Sea and in the Mediterranean, and least towards
the poles. In not a few parts of the globe we find that
currents bringing cold water of relatively low salinity
meet warm currents of highly saline water. When this
occurs the warmer water floats on the surface of the
colder, and there is a junction layer where a sudden
change in temperature, salinity, and specific gravity
occurs. Such a meeting of warm and cold currents is
found, for instance, in the Norwegian Sea, where the
Polar drift meets the warm currents of that sea; it
occurs also off the eastern coast of North America,
where the Labrador current meets the Gulf Stream ;
it occurs off the Cape of Good Hope, where the cold
Benguela current meets the warm Agulhas one, and
soon. Now, as has been already suggested, such regions

ANIMAL LIFE IN THE SEA 171

are regions of great mortality among the pelagic
organisms, for the stenothermal forms are killed by the
sudden changes of temperature which occur. But there
is something more than this. The delicate organisms
which float in the surface waters of the sea are neces-
sarily very sensitive to changes of density. They must
be able to float without effort—that is, their specific
gravity must be that of the water in which they float.
Very many possess certain powers of adjustment—that
is, they can vary their specific gravity in harmony with
variations in the specific gravity of the water. But
this power of adjustment has its limits. When warm,
light water is floating on cold, dense water, it is appar-
ently impossible for the more delicate forms at least to
pass through the junction layer of the two, which thus
forms what Dr. Hjort calls a ‘ false bottom ’, a region
where living and dead plankton animals accumulate.
By a series of very ingenious investigations he has shown
that, in many parts of the ocean, not only does such a
junction layer between the surface and lower layers of
water occur, but that this layer is a region of great wealth
of life. Itis a false shore-line in mid-ocean, and Dr. Hjort
has shown that many of the littoral fish at certain
seasons leave the shore, and swim out into the open
ocean, following this line of change of density, where
- food accumulates—-where it is as abundant as it is
near the land. There seems some reason to believe
that such an accumulation of pelagic organisms in
an area of sudden change of density occurs in the
vicinity of all the great fishing regions of the world,
and thus, as it were, prolongs the plenty of the shore
out into the open, and helps to account for the abund-
ance of fish.

As to the actual temperature of ocean water, the

172 THE DISTRIBUTION OF

statements made above in regard to the temperature
of littoral waters hold good here also. An important
point, however, is that nowhere—not even under the
Equator—does the warmth of the surface layer extend
far downwards. In practically all parts of the ocean
the temperature at depths exceeding 1,000 fathoms is
permanently low, and varies but little throughout
the year.

The conditions as to light demand detailed considera-
tion. Very careful experiments show that all rays of
light are present at depths of 50 fathoms. At 300
fathoms the red and green rays have disappeared, but
blue and violet rays remain. Some of the violet rays
are still present at 500 fathoms, for photographic plates
are slightly affected here. At 900 fathoms they remain
absolutely unaffected after long exposure, and here we
must suppose that absolute darkness reigns.

These statements are true for middle latitudes, but
another point has also to be considered. Obviously,
the direction of the rays is not the same in all latitudes,
and therefore they will penetrate to different depths
in different latitudes. Dr. Hjort gives the following
figures in regard to the depths at which the same in-
tensity of light is found in different latitudes :

In 33° N. lat. at about 430 fathoms depth.
In 50° N. lat. at about 270 fathoms depth.
In 67° N. lat. at about 100 fathoms depth.

This is important, because, as we shall see directly,
the intensity of light greatly influences the bathy-
metrical (i.e. depth) distribution of marine animals.

The basal food-supply of marine animals—in the
open as off the shore—consists of algae, so that we
must add to the above consideration of the conditions
of life in the open some account of the pelagic algae.

ANIMAL LIFE IN THE gnaN 173

In the first place it is noticeable that the larger algae
are unrepresented in the open sea. The weed which
floats at the surface of the Sargasso Sea is an apparent
exception, but it is apparent only, for this weed is torn
off the shores, especially off the Bahamas, and though
it proliferates vegetatively in the open, and can live
there for a time, it is not a truly pelagic form. The
swarms of animals which live among the floating
masses of weed in the Sargasso Sea show what an
important addition to the food-supply of the ocean
the weed is.

The basal food-supply of the pelagic animals is,
therefore, the phytoplankton, the minute algae which
float at the surface. Necessarily these algae, as light-
demanding organisms, are limited to the surface layers
of the water, being especially abundant in the upper-
most 25 fathoms. They occur, though in diminished
quantity, down to 50 fathoms. Those found in greater
depths are resting stages or dead forms. The dead
forms, as we have already suggested, tend to accumu-
late, at least temporarily, at levels where a sudden
increase of density occurs. As regards quantity, the
water of the open ocean seems to be poorer in algae
than coastal waters, but, as has been already sug-
gested, currents may carry the shore forms out to the
ocean, and there ‘is reason to believe that then the fish
of the shore may follow them.

The next point to be considered is the distribution
of pelagic animals. There is a tendency among many
writers, at least tacitly, to confine the term * pelagic’ to
those forms which live sufficiently near the surface for
light to penetrate. There is, however, no real justifica-
tion for this use. An animal living at a depth of
2,000 fathoms is as truly pelagic as one living at

174 THE DISTRIBUTION OF

a depth of 50 fathoms, if it can be proved that neither
has any dependence on the proximity of a substratum.
But here, obviously, a difficulty occurs. A sponge with
an attaching stalk, taken in deep water, is obviously an
abyssal form, for it was attached to the bottom. But
if a net which has been dragged along the bottom at
a depth of say 2,000 fathoms proves, when examined,
to contain fish, how can we tell whether these fish were
truly abyssal forms, or were merely caught as the net
was hauled up? The difficulties in coming to a decision
are great, and until recently there has been a tendency
to assume that all fish obtained in deep water, and
having a bizarre form and colouring, large eyes or no
eyes, and so on, were abyssal fish, feeding on the
bottom. Dr. Hjort, however, believes that the recent
(1910) expedition of the Michael Sars, taken in con-
junction with some previous observations, has yielded
evidence which proves that many of these curious fish
are truly pelagic, living at or below the light limit. If
he is right, then pelagic animals may be divided into
at least three groups: (1) surface forms, most abundant
in the upper 100 fathoms or so; (2) mesopelagic or
intermediate forms, found about the light limit, which
varies with the latitude ; (3) bathypelagic forms, which
occur below the limit to which the rays of light pene-
trate, but yet not upon the sea-floor, of which they
are independent.

We must now consider some of the characters of
pelagic organisms, beginning with the plankton. Plank-
ton animals are those able to float in water without
exertion, and have usually flattened and expanded
bodies. Sometimes, as in the very abundant copepods,
a globule of oil makes the body light. At other times,
as in some protozoa, and such forms as the sunfish,

ANIMAL LIFE IN THE SEA 175

the lightness is due.to the’ presence of air or another
gas. In the surface forms the body is often transparent
and colourless, so that it is invisible as it floats in the
clear water, while in other cases it is blue or violet, so
that it will equally be invisible when seen from below
against the sky. That these peculiarities are adaptive
is suggested by the way in which they appear in un-
related forms. Thus while the shore annelids are
brightly coloured, Tomopteris, a pelagic form, is colour-
less, and has an expanded body. The same peculiari-
ties appear in Sagitta, a worm of uncertain affinities ;
in Phyllirh6e, a curiously modified and_ shell-less
gasteropod; in Salpa, a pelagic tunicate, and so on.
The surface jellyfish are blue or violet or almost colour-
less, while in deep water bright-red forms occur. Many
of the pelagic forms of the surface are phosphorescent.
We see this in the protozoon called Noctiluca, in the
beautiful tunicate called Pyrosoma, in some pelagic
crustacea, and so forth. Surface pelagic forms are so
numerous that no useful purpose would be served by
giving a list, but it may be interesting to note those
groups which have no pelagic representatives. While,
as already suggested; very many types of Coelentera are
pelagic, there are no pelagic alcyonarians (sea-pens,
&c.), and there are no pelagic sponges. With one
exception, to be considered later, there are no pelagic
echinoderms, though their larvae are often pelagic.
There are no pelagic bivalve molluscs, though many
different kinds of gasteropods are adapted to life in
the open sea, the sea-butterflies (Pteropods) being good
examples. Some other less familiar groups are also
unrepresented in the open sea.

On the other hand, there are enormous numbers of
pelagic larvae. As young animals have often delicate

176 THE DISTRIBUTION OF

translucent bodies it would seem as though they could
be very readily adapted to the pelagic life. As two
interesting forms mention may be made of the trans-
parent larva of the eel, called Leptocephalus, because
its relation to the parent was for long unknown, and the
glass crab (Phyllosoma), which is the larva of the rock
lobster, but was similarly for long regarded as an
independent organism.

Very remarkable is the presence in the open sea of
the members of an insect genus, Halobates, a kind of
bug, related to the forms which skim over the surface
of fresh-water ponds. Some fifteen species of Halobates
are known from the warmer parts of the oceans. The
insects appear to be perfectly pelagic in that even the
eggs are produced in the open, but they may also
approach shores, for what reason is not known. The
female carries the eggs for a time attached to her own
body, and the young have been found on floating solid
bodies in the sea. Wings are completely absent, and
the body is covered with a greyish pubescence, which
shines in the sunlight. The animals breathe air, and
belong rather to the surface of the ocean than to its
waters.

In contrast with these plankton or drifting forms,
we have in the open sea a number of powerful swimmers
(nekton) able to make headway against the currents.
The most important of these are the whales and their
allies, whose distribution is tied to that of the animals
upon which they prey. Thus the huge cachalot or sperm
whale (Physeter macrocephalus) is found especially in
warm seas, where it finds the large cuttles upon which it
feeds. The related bottle-nose (Hyperoodon), which also
lives on cuttles, occurs in colder seas, and is found
especially in the North Atlantic. The whalebone whales |

ANIMAL LIFE IN THE SEA 177

(Balaena, &c.) feed on the floating plankton, especially
on sea-butterflies (Pteropods), and are very common
in cool and cold waters, though widely distributed. On
the other hand, the fish-eating dolphins and porpoises
especially frequent shallow waters, where the fish on
which they prey are most frequent.

With the exceptions already given, purely pelagic
fish are not very abundant, at least in the surface
waters. The reason is apparently that food near the
surface is not sufficiently abundant. Professor Bouvier
notes that the Princesse Alice, the Prince of Monaco’s
yacht, while cruising in mid-Atlantic, found practically
no surface fish, except when some floating wreckage,
with a burden of attached invertebrates, such as
crustacea, &c., drew them to the surface. Among the
surface forms may be noted the flying-fish (Exocoetus),
the beautiful ‘dolphin’ (Coryphaena), and its allies,
the bonito (Thynnus pelamys), another flying form,
related to the mackerel, which is also pelagic. All these
forms, however, despite their obvious adaptations to
pelagic life, are more abundant as the land is approached
than far away from land.

The above forms may be regarded as characteristic
of the upper 100 fathoms or so of water. We have next
to consider the mesoplankton and mesonekton—that is,
the forms which occur between the surface layers of
water and the bottom, near the light limit. The study
of these is necessarily a difficult matter, involving
the use of complicated methods not hitherto used at
very great depths. Provisionally, however, the follow-
ing statements may be made, these being based especi-
ally on the results obtained by the Michael Sars
expedition.

At depths between 200 to 300 fathoms, but varying

1404 M

178 THE DISTRIBUTION OF

with the latitude for the reasons already discussed,
there seems to be a fauna of silvery or iridescent
animals, of which the fish have been chiefly investi-
gated, though medusae, and doubtless other forms also
occur. The fish living at these depths inhabit an
environment where a considerable amount of sunlight
penetrates, especially blue and violet rays. They have
often large eyes, which sometimes protrude at the end
of short stalks, and are then called telescopic. Large
light organs are present, which perhaps catch the last
rays of sunlight. An example is the silvery fish Argyro-
pelecus affinis, which lives in the Atlantic at depths of
somewhere about 200 fathoms; at night, when the
light conditions are the same near the surface as in the
depths where it habitually dwells, this fish has been
found within some 80 fathoms of the surface. Its
young have the same colouring as their parents, and
live in the same depths. Another form with the same
colouring and characters is Cyclothone signata, which
lives at a depth of about 300 fathoms.

Still deeper, at depths of from 300 to 500 fathoms,
where all trace of red and green rays have completely
disappeared, occur forms in which the colouring is
black or red. In the absence of any red rays both
these colours must look the same, and either will pro-
tect the animal. At these depths occur deep-red or
chocolate-coloured medusae, the only pelagic echino-
derm (Pelagothuria), bright-red prawns, such as
Acanthephyra multispina, and red or black fish, Se-
bastes norvegicus being an example of the former and
Goniostoma elongatum of the latter. The last-named
fish is purplish-black, with small light organs and
small eyes. It has huge jaws, a common feature in
these deep-water fish. A very curious point about it

ANIMAL LIFE IN THE SEA 179

is that in the daytime it lives at about 300 fathoms,
but at night it rises to within 80 fathoms of the surface.
Its young are crystal clear, and are surface forms living
in the upper 80 fathoms. On the other hand, some
members of the genus Alepocephalus live as far down
as 1,100 fathoms, and have no light organs. Their
young are the same black colour as the parents, and
live also in the great depths.

To sum up, so far as we know at present, forms
which live in the surface waters of the ocean are crystal
clear or pale violet or blue. Those which live in layers
where only blue and violet rays penetrate freely tend
to be silvery and to have large eyes and large light
organs. Still deeper, where no red and no green rays,
even in some instances no rays at all, penetrate, the
colour becomes red or black, and light organs are small
or disappear. That there is an intimate relation between
the characters and colouring of the organism and the
amount of light present is shown by the fact that when
the larva inhabits the same depth as the parent it has
the same colours. When it inhabits another zone it
takes on the normal colouring of the inhabitants of
that zone. Further, the fact that the forms from
deeper water occur in different layers of water in
different latitudes—the level corresponding to the
degree of penetration of light in the particular latitude,
so that the animals live higher up in high latitudes
than in low—suggests that the life of the animals is
strongly influenced by the amount of light present.

One other point is interesting in regard to the life
of these depths. Just as the littoral fishes often send
pelagic larvae out to the open sea, so some of the
abyssal fishes seem to have bathypelagic larvae, i.e.
larvae living in the open water, but at considerable

M 2

180 THE DISTRIBUTION OF

depths. Thus the members of the genus Macrurus,
believed to be true abyssal, that is ground-living, forms,
have apparently such eggs and larvae. The abyssal
hermit crabs, also, have bathypelagic larvae, just as
the littoral forms have larvae which form part of the
surface plankton.

(3) The abyssal region of the oceans is contrasted
with both the other regions in that its waters must
be permanently calm. It is in harmony with this
calm that we find, as already stated, some deep-sea
sponges which have no attachment to the substratum,
upon which they seem to lie loosely. The pressure
must be enormous, for the animals of the depths sup-
port a weight of several miles of sea-water upon their
bodies. Though there are no plants, oxygen seems to
be abundant. But this is a consequence of the well-
developed circulation of the great oceans, and is not
a feature of enclosed seas. Thus the Black Sea has no
animals in its depths, and the waters there are impreg-
nated with sulphuretted hydrogen, and contain but
little oxygen. In the Mediterranean the conditions are
less extreme, but the deep-sea fauna is scanty. Light
is, of course, entirely absent in the depths. The water
is permanently cold. The basal food-supply must
consist of the dead animals of the upper zone, which
fall downwards through the water as they die. In
consequence we find that many of the abyssal forms
are mud-eaters. To what extent the floor of the
ocean is peopled still remains a difficult question.
As already noted, Dr. Hjort believes that genuine
-abyssal fish, in the sense of ground forms, are few,
most of the so-called abyssal forms being bathypelagic
and having been obtained in the process of hauling
in the dredge. But there can be no doubt that many

ANIMAL LIFE IN THE SEA 181

invertebrates do occur on the sea-floor in the greatest
depths, especially over regions covered by Globigerina
ooze. Among these invertebrates sponges and echino-
derms preponderate, molluscs and crabs among crus-
tacea being rare. Of the sponges the old-fashioned
siliceous forms are numerous, and calcareous sponges
are absent. A few corals occur. Among the echino-
derms all the classes are represented, and the old-
fashioned crinoids, rare in shallow water, are relatively
abundant. Among the sea-urchins the special feature
is the abundance of old-fashioned irregular forms, which
occur so abundantly as fossils. Annelids occur especi-
ally in the red clay. In harmony with the great
uniformity of the conditions over wide areas, the
abyssal animals are widely distributed. Further, as
the conditions of temperature, and during the long
Arctic night the conditions regarding illumination,
show resemblance in polar and abyssal regions, there is
apparently some resemblance between polar animals
and abyssal animals.

RerFerences. The forty-two volumes of the ‘Challenger’ Reports (1880-
91) give the results of the epoch-making cruise of that vessel. More
popular accounts of its work are to be found in Mosely, Notes by a Natural-
ist on the ‘ Challenger’ (London, 2nd ed., 1892), and Wyville Thomson,
The Depths of the Sea (London, 1873), and Narrative of the Voyage of the
‘Challenger’ (London, 1885). The following may also be consulted :
Ortmann, Grundziige der marinen Tiergeographie (Jena, 1896); Chun,
Aus den Tiefen des Welimeeres (Jena, 1900); Marshall, Die Tiefsee und
thr Leben. For the more recent work discussed in the present chapter,
see Hjort, The ‘ Michael Sars’ North Atlantic Deep-sea Expedition (Geo-
graphical Journal, April and May, 1911); Hjort, Die erste Nordmeer-
fahrt des norwegischen Fischdampfers ‘ Michael Sars’ im Jahre 1900
(Peter ’s Mitteilungen, IV, 1901), and The Depths of the Ocean, by
Sir John Murray and Dr. Johann Hjort (London, 1912). Two interesting
popular articles on the work of the Princesse Alice are Les Vertébrés de
la Surface and La Faune Pélagique des Invertébrés, by Professor Bouvier
(Revue Générale des Sciences, 1906).

CHAPTER IX

THE ANIMALS OF LAKES AND RIVERS

Tuer water of rivers and of very many lakes differs
from that of the sea in the absence of any considerable
quantity of common salt, making it ‘fresh’ instead
of salt. Such bodies of water contain a special fauna,
which, as contrasted with the fauna of the ocean, may
be described as impoverished ; but in addition to this
negative character, this fauna possesses also some
positive ones, which we shall consider later. Certain
other lakes, either because they had a former connexion
with the sea, or because they have no outlet, contain
saline water. Such masses of water may contain some
animals with distinctly marine affinities, e.g. in the
Caspian are found a seal and a kind of herring. But
the Caspian is far from being as salt as the sea, and in
addition to animals of marine type it contains some
typically fresh-water fishes. On the other hand, masses
of very salt water, such as the Great Salt Lake of Utah,
contain peculiar brine-shrimps (Artemia) not found
in the sea. A still further complication is introduced
by the fact that certain lakes, though their waters ate
perfectly fresh, contain animals of distinctly ‘ marine ’
facies. Thus Lake Baikal, though its waters are not
salt, lodges a seal allied to the Caspian seal, and a
marine worm; Lake Tanganyika, in Central Africa,
contains several animals of marine type, and so on.
In consequence we cannot sharply separate fresh-
water animals from marine ones, in the sense of imply-

THE ANIMALS OF LAKES AND RIVERS 183

ing that all the latter live only in the ocean, and that
lakes and rivers contain only the former; nor can we
assume that all aquatic animals can be divided into
marine and fresh-water types, for this would exclude
forms like the brine-shrimps, characteristic of salt
lakes and brine pools. But, at the same time, it is
convenient to recognize a general distinction between
the animals of the oceans, the marine forms, and those
characteristic of lakes and rivers, which are for the
most part fresh-water animals.

Using the terms in this general sense, we may note
that the poverty of the fresh-water fauna is due to the
relatively unfavourable conditions. Among these un-
favourable conditions we probably need not include
the absence of considerable amounts of salt dissolved
in the water, for the presence of an appreciable amount
of salt, e.g. in the Caspian, does not give rise to a
marked increase in the fauna as compared with fresh-
water lakes, and the presence of a large amount of
salt, e.g. in the Great Salt Lake, is distinctly inimical
to animal life.

Before proceeding to discuss the especially unfavour-
able conditions, we may note that it is generally
believed that life originated in the sea, not in fresh
water, and that therefore all the animals now found
in rivers and lakes must have originated in one of two
ways: (1) they may have been derived directly from
marine forms, or (2) they may have been derived from
land animals which have reacquired the aquatic habit
of their remote ancestors.

Of the first group, fresh-water fishes may be taken
as examples. They have certainly had marine ances-
tors, and in some cases, e.g. the eel and the salmon,
the existing forms have retained the power of living in

184 THE ANIMALS OF LAKES AND RIVERS

the sea as well as in fresh water. Similarly, the bivalve
molluscs of the lakes and streams have certainly had
marine ancestors. Most of the higher crustacea live in
the sea, and the few crayfish and prawns which live
in fresh water are certainly descended from ancestors
which lived in the open ocean.

It is otherwise with most of those numerous forms
which, though they live in water, are yet adapted for
breathing air. The otter which haunts the salmon
rivers has lungs no less than its distant relative the
polecat ; the water-tortoises have lungs like the land
forms ; the water-beetles and the many insect larvae
found in pools, the water-spider, the water-mites, the
pond-snails, are all air-breathers, and their peculiar
respiratory organs would be inexplicable if we could
not assume that their immediate ancestors lived on
land. In these cases there is no reason to believe that
the animals have ever sought the sea since the time
when their far-off ancestors acquired terrestrial charac-
ters. Therefore in possessing some of these forms
the fresh-water fauna is not poorer but richer than
the sea, which has hardly any insects (cf. p. 176), no
air-breathing molluscs, no true spiders, and so on.
Again, while: most amphibians have a fresh-water
larval stage, and some pass much of their life in water,
no living amphibian is ever found at any period of its
life in the sea.

But when we come to consider the groups which are
richly represented in the sea, the tale is different. The
echinoderms are exclusively marine—no starfish, sea-
urchin, brittle-star, nor sea-lily is known from fresh
water. No cuttle hunts its prey in the depths of the
great lakes, nor creeps along river bottoms. The poly-
chaete worms are almost exclusively marine, and the

THE ANIMALS OF LAKES AND RIVERS 185

higher crustacea largely so. The great group of the
Coelentera is very scantily represented in fresh water,
where also there are but few sponges. Fresh-water fish
also are few in number as compared with their marine
allies, though they include some peculiar forms, like
the ganoids (e.g. sturgeon, bony pike, &c.), and the
dipnoi or double-breathers. It is, however, interesting
to note that of these peculiar fish all the three living
dipnoi, and at least one ganoid (Polypterus) have
accessory breathing organs which enable them to breathe
air, as well as gills. This fact, taken in conjunction
with the relatively large numbers of air-breathing
invertebrates, suggests that one difficulty in colonizing
fresh water has always been its relative deficiency in
oxygen. In the general case this applies, of course, to
lakes and inland seas rather than to rivers, whose
water is oxygenated by its motion. But where a
marked dry season occurs, as in many parts of the
globe, the rivers may periodically cease to flow, and
be represented by a series of stagnant pools, whose
waters become very fetid—conditions highly unfavour-
able to animal life.

In the lakes of temperate regions, where the surface-
water periodically cools below the point of maximum
density of water, a vertical circulation is produced which
carries down oxygen to the deeper layers of the water,
at least at certain seasons. This makes life in the
deeper layers possible, though some American experi-
ments suggest that the oxygenation of these deeper
layers occurs only in spring and autumn, as the ice
melts in the former case, and before it is formed in
the latter. In the summer the heating of the surface-
water gives rise to a stratification into two layers,
when the warm oxygenated surface-water lies upon a

186 THE ANIMALS OF LAKES AND RIVERS

deeper, colder layer, whose oxygen is only renewed
very slowly. This relative paucity of oxygen naturally
checks life in the deeper parts of the lake.

In tropical lakes the fact that the surface-water does
not cool in the same way in autumn, and that there is
no spring melting of ice, makes the vertical circulation
much less marked, and leads to a virtual absence of
a deep-water fauna. In brief, then, the deep-water
fauna of lakes is always poor, because of the absence
of a well-marked circulation of the water, and the defi-
ciency in the amount of oxygen, with the resultant
paucity of life, increases towards the Equator.

The second unfavourable condition in fresh water is
its frequent turbidity. Rivers are powerful eroding
and transporting agents, lakes are great filter-beds on
the course of rivers; necessarily, therefore, the waters
of both must often contain fine particles in suspension,
which would form a deposit on the surface of the
breathing organs of animals living in them, and greatly
retard respiration. The importance of this may be
illustrated by an example. The Firth of Forth is
a great estuary swept by the tides, once inhabited by
great numbers of marine animals tolerant of estuarine
conditions, notably by oysters. There is much evidence
to show that its waters are becoming muddier year by
year, apparently because drainage of the ‘ mosses’ in
the upper regions of the river has made the run-off
there more rapid. As a result of the increasing muddi-
ness many animals have disappeared—the oyster-beds
long ago ceased to be productive, and other forms also
are diminishing. Now if an increase in the amount of
mud in the water kills estuarine animals, which are
naturally tolerant of some suspended matter, even
a small amount of suspended matter will prevent the

THE ANIMALS OF LAKES AND RIVERS 187

growth of the more delicate animals, and must be
a great bar to the passage of marine animals up estu-
aries and rivers. Almost any coast-line will illustrate
the same thing on a smaller scale. The shore collector
knows that to get the rarer and more delicate animals
he must seek rocks or beaches remote from the mouths
of streams, for at the immediate outlets of these only
the hardiest forms occur. This statement is not in-
consistent with that already made, that the waste of
the land is an important part of the food of the sea-
animals, for oxygen is a prime necessity which precedes
even the need of food, and few animals can take in
oxygen if their respiratory organs are choked with mud.

But it is not only the suspended matter in the water
of lakes and rivers which is inimical to animal life.
The presence of large amounts of lime or manganese
salts, or of humic acid, may render the water unsuitable
to certain forms.

Still another point of great importance is the enor-
mous variation in temperature to which lakes and
rivers are subject. These variations are normally so
great that they render the water absolutely unsuited
to stenothermal animals. All fresh-water forms must
be eurythermal (cf. p. 160), and temperate and polar
forms must have special means of protecting them-
selves, or their offspring, against the cold of winter.
In the ‘lakes’ of Victoria Land, which are almost per-
manently frozen, the naturalists of Sir Ernest Shackle-
ton’s expedition found rotifers which can apparently
tolerate being frozen into the ice for years, and will
yet hatch out if the temperature rises and the ice melts.
This is an extreme case, but it suggests the adaptations
necessary before animals can thrive in small masses of
water. Low temperatures are not, however, the only

188 THE ANIMALS OF LAKES AND RIVERS

danger. High temperatures, with resultant desicca-
tion, are just as dangerous, and must also be provided
against. The amoebae in a waterspout can be dried to
dust, and will lie unharmed within their protective
cysts until the rain comes, and the spout once again
becomes suitable for their active life. In the swamps
of the Paraguay or of the Central African rivers, the
lung-fishes lie in their mud cases, waiting till the wet
season again permits them to become active. Drought
and freezing, high temperatures and low, are risks
which every fresh-water organism must face, and the
lessons learnt in the conflict with them are stamped
deep on all the inhabitants of stream and lake. In
these, as on dry land, life in the temperate and frigid
zones is markedly seasonal, and it is interesting to note
that where the problem of seasonal adaptation cannot
be directly faced, it may be avoided. For example,
those shallow bodies of water which are favoured by
frogs as spawning-places are very apt to dry up in full
summer, but the tadpole avoids the risk of drought
by adjusting its time of metamorphosis to the time of
drought, so that it is ready to leave the water at the
time when the pond would naturally dry up. Every
summer one may find cases where the adjustment has
not been sufficiently delicate, so that thousands of tad-
poles die because the drought comes before they are
ready for their land life. In autumn the ponds fill again,
and the little frogs return again to them to pass the
winter in their mud.

One other unfavourable feature of rivers and lakes
is found in the strength of their currents. The waters
of the ocean are swayed also by currents, but within
considerable limits, and with the exceptions already
given (p. 161), such currents cause little harm to

THE ANIMALS OF LAKES AND RIVERS 189

marine animals. Tides and currents sweep the members
of the plankton hither and thither, but the regions
into which they are swept may be more suitable than
those they have left—in the case of the littoral animals
the currents are the great distributors. But for a lake
or river animal to be swept out to sea must in the
general case mean death, for it means the passage from
a specialized environment to which the animal has
become adapted through progressive variation, to
another to which it has no adaptations.

In the nature of things littoral animals must have
always had special facilities—that is, facilities in regard
to space—for conquering fresh water. So far as their
relations in space go, it is relatively easy for them to
progress, actively or passively, from the shore up
rivers and estuaries. But we have already seen that
it is one of the great peculiarities of littoral animals
that they tend to produce free-swimming young, or
free-swimming stages, whose purpose it is to ensure
distribution along the shore, and the existence of these
free-swimming stages or young offers a serious obstacle
to the colonization of fresh water. For example, in
some artificial ponds at the mouth of the Ganges
a very curious fauna has been described (Records Indian
Museum, i. (1907) p. 35). These ponds are sometimes
connected with the sea, and sometimes shut off from
it, and are sometimes filled with water rendered very
salt by evaporation, while at the rainy season the water
is nearly fresh. In spite of these variations in salinity,
a number of marine forms occur, including a sea-
anemone, a kind of sea-fir (Hydromedusan), and a
polychaete worm, showing that it is not its freshness
alone which prevents such forms from colonizing fresh
water. Now sea-anemones produce usually free-

190 THE ANIMALS OF LAKES AND RIVERS

swimming larvae called planulae ; Hydromedusae
produce free-swimming medusoids ; polychaete worms
have free-swimming larvae called trochospheres, all
delicate forms incapable of swimming against currents.
These particular ponds have at intervals a connexion
with the sea, and so can be peopled afresh from it, but
if we suppose that at some future period the connexion
with the sea were to be reduced to a swift-flowing
stream, could we suppose that the animals would
persist? Obviously not, for the medusoids, the
planulae and the trochospheres would tend to be swept
out to sea by the stream, and there would be thus no
young forms to replace the parents when death took
place. So real is this danger that, with rare excep-
tions, fresh-water organisms have no free-swimming
stage in their life-history. Only those marine animals
could colonize fresh water which had no plankton
stage in their life-history, or which managed somehow
to survive until this free-living stage could be sup-
pressed by progressive variation. This perhaps gives
us a second reason why so many animals with a
terrestrial ancestry have succeeded in colonizing fresh
water, for these forms would not have a free-swimming
stage. Thus while most marine gasteropods have free-
swimming larvae, fresh-water snails, like land snails,
hatch as miniature adults, which makes life in ponds
and streams easy for them. These arguments of course
only apply to those delicate organisms which are in-
capable of swimming against currents. Relatively
powerful swimmers, like many insect larvae, can
resist moderate currents, though they avoid regions
where the water is in rapid movement.

But, it may be said, in a large lake the need for dis-
tribution is as great as in the sea, how do. the lake

THE ANIMALS OF LAKES AND RIVERS 191

animals succeed in spreading from one part of the lake
to another if they have no free-swimming stage? The
answer is apparently that the distribution is largely
passive, being assisted by sudden freshets, which may
produce currents of unusual violence, capable of carry-
ing adults, or, more likely, resting eggs to a distance.
In addition there are some curious special adaptations.
We may take two examples of familiar forms. The
fresh-water hydra as contrasted with the marine sea-
firs has no free-swimming stage. While the weather is
warm it reproduces itself by buds; towards autumn
it produces coated eggs which survive the winter, and
produce fresh hydrae in spring. These shelled eggs
are doubtless easily transported by currents. More
remarkable are the conditions in the fresh-water
mussel (Anodon). Here the mother retains the young
within her body till sticklebacks approach the mud in
which she is living. She then discharges the young,
who fix themselves to the fish, and are thus carried
about for a time, and so distributed. Later they drop
from the stickleback to the mud, and develop into
mussels like the parent.

With this preliminary account of the conditions of
life in lakes and rivers, and of the resultant adapta-
tions, we may proceed to consider the animals which
constitute the fauna.

A considerable number of mammals show at least
a partial adaptation to fresh-water life. Thus we have
many kinds of otters, which have webbed feet and feed
upon fish ; not a few insectivores, such as the common
water-shrew and the Russian desman ; various rodents,
such as the beaver and the water-vole ; a monotreme,
the curious duck-mole or Ornithorhynchus of Australia,
and soon. In most of these cases the modifications are

192 THE ANIMALS OF LAKES AND RIVERS

not very profound, not sufficient to cause the animals
to diverge very markedly from their terrestrial allies.
This suggests that in most cases the modification is
relatively recent. Among the more frequent modifica-
tions are webbed feet, used in swimming, close dense
fur which prevents the water reaching the skin and so
chilling it, and a broad flattened tail used as a rudder.
Of mammals with more profound adaptations to
aquatic life, few occur in inland waters. We have
already spoken of the Lake Baikal seal (Phoca sibirica)
and the Caspian seal (Phoca caspica), both closely
related to the Ringed seal of northern waters. The
Baikal seal also occurs in Lake Oron, on the course of
the river Vitim, and in addition to its seal Lake Baikal
shows certain ‘marine’ features in its molluscan fauna.
The explanation is perhaps that the lake had once
a connexion, direct or indirect, with the great inland
sea of Tertiary times. Among the Sirenia the manatee
is largely fresh-water, ascending such rivers as the
Amazon almost to their sources. But these animals
thrive equally well in the ocean, and their presence in
large rivers requires no special explanation.

The large Cetaceans are exclusively marine, but
certain dolphins occur in fresh water. The one which
is most definitely adapted for this habitat is the blind
dolphin of the Ganges, Brahmaputra, and Indus
(Platanista gangetica). This animal occurs from tidal
waters to the upper reaches of the rivers, as high up
as the depth of the water permits, but it has never
been seen in the sea. These dolphins feed upon fresh-
water fish and river prawns, and are apparently primi-
tive forms which, like Ganoid fish, have found a refuge
in fresh water. Other fresh-water dolphins occur in the
large rivers of South America, especially the Amazon.

THE ANIMALS OF LAKES AND RIVERS 193

A very large number of birds haunt the vicinity of
fresh water, being attracted to it, especially at certain
seasons, by the relative abundance of food there.
Such forms, however, live on the surface rather than
in the water.

In regard to reptiles it is interesting to note that
among the Chelonia, as among mammals, the adapta-
tions to the aquatic life are much more marked in
marine than in fresh-water forms. Marine turtles have
their limbs turned into paddles ; fresh-water Chelonians
swim by limbs which are only slightly modified as com-
pared with terrestrial forms. This is quite parallel to
the conditions which occur in mammals. Such typically
marine mammals as seals, Cetaceans, and Sirenians
have their limbs (when present) turned into paddles.
Such fresh-water carnivores as otters, fresh-water
rodents like water-voles and beavers, fresh-water in-
sectivores like water-shrews, have only slightly modified
limbs, retaining many features found in terrestrial
types. This is equivalent to saying that for an air-
breathing animal the modifications which fit an animal
for life in the sea must necessarily be more profound
than those which fit it for life in fresh water, which
usually occurs in relatively small masses, and where
the animal inhabitants often retain considerable depen-
dence upon the land.

Among the Chelonia a considerable number spend
at least a great part of their time in the water. As
two examples we may take the American Snapping
Turtle (Chelydra serpentina), a form reaching a con-
siderable size, which is found in lakes, ponds, and rivers
throughout the greater part of North America. It has
strongly webbed feet and is carnivorous. As might be
deduced from its limbs, which are but little modified,

1404 N

194 THE ANIMALS OF LAKES AND RIVERS

it leaves the water, should want of food or any
other cause render this necessary, and is capable of
(awkward) progression on land, in a way that would
be impossible for a marine turtle. In Central and
Southern Europe the ponds and swamps are inhabited
by the pretty little pond tortoise (Zmys europea), which
has also webbed feet. These little creatures are fond
of coming out of the water to bask, and migrate from
one pond to another if the first shows signs of drying up.

The same tale of partial adaptation to aquatic life
can be told of the various kinds of crocodiles, which,
like the forms just mentioned, seem to seek the water
for the sake of the food to be obtained there. The
living crocodiles are still so far bound to the land that
they leave the water voluntarily, either to bask in the
sun, or in order to migrate from one region to another,
or to lay their eggs.

None of the snakes show such definite adaptations
to life in fresh water as do the sea-snakes to life in the
sea, but certain forms enter the water freely in search
-of food. This is especially true of the common grass-
snake (Tropidonotus natrix), which sometimes spends
the greater part of its life in water, in which it swims
easily.

The amphibians show some very curious conditions
in connexion with adaptations to the aquatic life. The
normal condition, as already suggested, is that the
young live in the water and breathe by gills, while the
adults live on land and breathe by lungs. But there
are some interesting exceptions. Throughout much of
North America, and especially in Mexico, there occurs
a salamander (Amblystoma tigrinum), which when adult
is normally terrestrial, except at the breeding-season,
and is of a brownish colour, with yellow spots and

THE ANIMALS OF LAKES AND RIVERS 195

blotches. The larvae, known as axolotls, have large
external gills, slender limbs, and a tail fringed with
a swimming membrane. Now, at times, under natural
conditions, but for causes not clearly understood, these
larvae may become sexually mature and lay eggs, while
remaining in the water, and while retaining larval
characters. That is, of the Amblystoma eggs laid in
a lake some may give rise to the salamander-like adult,
while others may breed while retaining tadpole charac-
ters. As a much rarer phenomenon this may occur
in the common newts, which sometimes breed while
retaining the characters of the tadpole.

In a more advanced form the same condition occurs
in certain other amphibia, which, as it were, never grow
up, but remain permanently in the tadpole stage. In
the lakes and swamps of the eastern part of the Missis-
sippi basin, and in some of the Canadian lakes, there
occurs an amphibian of about one foot in length, with
three pairs of external gills and a fringed tail, called
Necturus maculatus. This animal uses both its lungs
and its gills for breathing purposes, and retains its gills
throughout life. The same statements may be made of
Siren lacertina, the mud-eel of parts of the United
States. These are examples of the reacquisition
of the exclusively aquatic habit by animals whose
ancestors had become adapted for life both on land and
in water. Amphibians, it may be noted, have probably
been derived from fishes of the ‘ ganoid’ type—that is,
originated from fresh-water forms (see p. 202).

This brief account of the air-breathing vertebrates
found in lakes and rivers enables us to draw some
general conclusions in regard to these. Obviously we
can classify them under three headings: (1) The
amphibians, which possess gills at some stage in their

N2

196 THE ANIMALS OF LAKES AND RIVERS

life-history as well as lungs, illustrate the transition
from the fresh-water to the terrestrial life, some forms
showing a tendency to revert to the aquatic life ;
(2) such forms as otter and desman among mammals,
water-tortoise and crocodile among reptiles, show land
animals in the process of reacquiring aquatic charac-
ters, either because food is easier to get in water than
on land,. or because life is safer there for relatively
helpless animals; (3) the fresh-water dolphin of the
Ganges, the manatee, &c., exemplify the acquisition
of the fresh-water habit by animals adapted to life in
the sea, but driven from the sea by the competition of
more advanced forms, or quitting it voluntarily in
search of food. The seals found in Lake Baikal and the
Caspian Sea are truly marine forms, more or less
accidentally cut off from their natural habitat.

The remaining animals of lakes and rivers, includ-
ing fishes and invertebrates, are of more ancient origin,
and have had more time to become fundamentally
modified.. The manatee is still a marine animal, the
fresh-water dolphin of the Ganges became fluviatile at
a period which is geologically but of yesterday, but the
fresh-water hydra, the fresh-water crayfish, the fresh-
water mussels have had time to become greatly modified.

The fauna of swift rivers must in the general case
consist only of powerful swimmers like fish and fresh-
water crayfish, or of animals which can so fix them-
selves as to avoid the force of the currents. In lakes
and ponds, on the other hand, there is a greater wealth
of life, and it is possible, as in the sea, to divide this
life into littoral, pelagic, and abyssal groups. But of
these groups the littoral is by far the largest, because
it is enriched by many forms, like the larvae of insects,
which retain some dependence upon the land. The

THE ANIMALS OF LAKES AND RIVERS 197

abyssal fauna is sometimes virtually absent, e.g. in the
Caspian, owing to the paucity of oxygen, and its members
rarely if ever show the specialization seen in the
animals of the ocean abysses. This seems to be partly
because lakes are necessarily temporary phenomena,
destined to be filled up after a longer or shorter period,
and therefore there is no time for a special abyssal
fauna to develop. As considerable differences exist in
the fauna of different lakes, it seems better to give
some account of a few types rather than to make
general statements in regard to lakes as a whole.

The fauna of the lochs of Scotland has been studied
in great detail by the members of the Scottish Loch
Survey, so that it is possible to make a considerable
number of general statements in regard to it.

The members of the Survey studied in all 562 lochs,
and, exclusive of vertebrates and of insects and their
larvae, 440 species of animals were found. Most of the
lochs studied are shallow, but Loch Morar has a maxi-
mum depth of over 1,000 feet, and Loch Ness of over
750 feet, rendering an abyssal fauna possible. The
members of the Survey did not especially investigate
the fishes. We know, however, from other sources
that the special feature is the number of members of
the salmon family, which run into many species or
varieties; cf. the Loch Leven trout, which is peculiar
to that lake.

Of the invertebrates the majority may be described
as littoral in that they occur in shallow water, near
the margins of the lochs. Here are found many insect
larvae, and not a few adult insects adapted for life in
the water, e.g. water-beetles and water-boatmen. Here,
too, are found nearly all the few molluscs, including the
large mussels and the small fresh-water snails. Small

198 THE ANIMALS OF LAKES AND RIVERS

crustacea are numerous, especially those forms popu-
larly called water-fleas, from their jumping movements.
Many rotifers occur, together with such small animals
as water-mites, water-bears (Tardigrada), and so forth.

Only some thirty species are typically pelagic and
thus constitute the plankton. These comprise fourteen
small crustacea, twelve rotifers, and four protozoa.
Of these thirty species most are very widely distributed,
the animals which constitute the fresh-water plankton
being all but cosmopolitan, and believed to form ‘ the
oldest community of organisms on the earth ’ (Sir John
Murray). The special feature, however, is the presence
here of some Arctic crustacea, and the absence of certain
forms found in other European lakes, both no doubt
due to the northern position of the lochs.

As has been just stated, the lochs are mostly shallow,
and therefore an abyssal fauna is rarely developed.
Of the two deep lochs, only in Loch Ness were successful
deep-water dredgings made. Here it was found that at
depths greater than 300 feet the majority of the littoral
species disappear, leaving a small group of animals,
including one mollusc (a small hivalve called Pisidium
pusillum), three crustacea, three worms, an insect larva,
and a few infusoria. These occur also in the littoral
region, and the forms from deep water show no special
feature, so that it would be more correct to say that
a few only of the littoral forms can live in depths
greater than 300 feet than to say that a special deep-
water fauna exists.

Before leaving the fauna of these lakes, one interest-
ing point may be noticed. In certain of them, especially
those which lie but little above sea-level, a marine
crustacean called the opossum shrimp (Mysis) occurs.
In the Scottish lochs the species of Mysis found is that

THE ANIMALS OF LAKES AND RIVERS 199

which also occurs round the coasts in salt and brackish
water, and its presence is doubtless to be explained on
the supposition that it is a recent migrant from the
sea. But in certain continental lakes, especially those
in Denmark and on the North German plain, another
variety or species occurs, called Mysis relicta. The
modifications which this form displays are believed to
indicate that the bodies of water in which it occurs
were once connected with the sea, but have been long
separated from it. Such organisms are said to con-
stitute a relict fauna, and the lakes in which they occur
are called relict lakes (German, Reliktenseen), to indi-
cate their supposed former connexion with the sea.

With the Scottish lochs may be contrasted Lake
Balaton or Platten See, a large shallow body of water
in Hungary, connected with the Danube drainage
system. The area of this lake is about 250 miles, but
its average depth is only 10 feet, and the maximum
36 feet. Its fauna has been subjected to detailed
investigation.

As its shallowness makes Lake Balaton a mere pool,
we find that here the fauna forms a unity. Necessarily
there is no deep-water fauna, and it is not possible to
discriminaté accurately between littoral and pelagic
faunas, as the bottom is everywhere so close to the
surface. The total number of species does not differ
greatly from that of the Scottish lochs, being about
475, including 38 kinds of fish, but exclusive of insects
and. air-breathing vertebrates. The fish show some.
interesting peculiarities. There are no trout, but the
lake contains such common fresh-water forms as perch,
gobies, pike, eels, roach, tench, gudgeon, &c., and in
addition two forms of special interest. One is the cat-
fish called ‘Wels’ by the Germans (Silurus glans),

200 THE ANIMALS OF LAKES AND RIVERS

and the other the sterlet (Acipenser ruthenus). The
cat-fish constitute a large family of fresh-water fish
especially characteristic of the equatorial region. The
wels is found only in Eastern Europe, and is the only
member of its family in European waters. The sterlet
is a near ally of the sturgeon, and is a member of
a small group of old-fashioned fishes which used to
be called ganoids. The name has been abandoned in
more recent classifications of fish, for the ganoids are
not a homogeneous group, but their interest for us is
that they are fish of a primitive type, which once lived
in the sea, but have been driven into fresh water
(though the sturgeon also occurs in the sea) by the
competition of the more highly organized bony fish.

Of the invertebrates it is sufficient to say that
with considerable general resemblance to those of the
‘Scottish lochs, they display certain minor differences.
Thus, the Arctic types of crustacea are absent, and
such forms as the fresh-water crayfish (Astacus), found
in England, but not in Scotland, are present here.
Similarly there is a greater wealth of molluscs, a con-
siderable number of genera of more or less southern
facies being here represented, though they are absent
from the lochs of Scotland. On the other hand, the
rotifers are far less abundant than in the Scottish lochs,
but comparisons of this sort do not profit very much,
as it is difficult to be sure that the investigations have
been conducted along exactly similar lines in the two
cases. F

As a third example of a lake fauna we may take that
of Lake Tanganyika in tropical Africa, which has been
the object of careful study by Mr. J. E. S. Moore
and others.

Tanganyika has an area of 12,700 square miles. It

THE ANIMALS OF LAKES AND RIVERS 201

has never been systematically sounded, but depths of
1,200 to 2,100 feet have been recorded at various
localities. The elevation is 2,624 feet above sea-level.

The fish fauna show some curious features. In the
first place the lake contains a species of Protopterus,
one of the three living genera of dipnoi or lung-fishes.
The two other living genera inhabit the one (Lepido-
siren) the rivers of tropical America, and the other
(Ceratodus) those of Queensland. As well as occurring
in Lake Tanganyika, the species of Protopterus have
a wide distribution in the lakes and rivers of the middle
portion of the African continent. All the dipnoi
possess, as we have seen, both lungs and gills, but the
two sets of breathing-organs are not in use simul-
taneously. The fish normally inhabit regions where a
periodical dry season occurs, when the water in which
they live either becomes very foul or dries up. The
first condition is frequent in the natural habitat of the
Queensland Ceratodus, the second in that of the two
other genera. At this period, therefore, Lepidosiren and
Protopterus make themselves nests of mud, in which
they lié dormant, breathing by their lungs till the
water returns. In the case of Ceratodus the lungs are
‘used when the water is too foul for the gills to be of
any use, but the animal does not make mud nests. All
three genera show certain very primitive characters
in combination with this specialized mode of breathing.
Their ancestors were certainly marine, many fossil
forms having been found. Thus the dipnoi must be
looked upon as the fresh-water survivors of a highly
primitive group of marine fishes, owing their per-
sistence to the acquisition of specialized breathing-
organs. Their ‘discontinuous distribution’ over the
surface of the globe is another point of interest in

202 THE ANIMALS OF LAKES AND RIVERS

regard to them. Such discontinuity is common with
old-fashioned animals, which owe their local persis-
tence to their adaptation to special environmental con-
ditions. The tapir in the swampy forests of the Malay
region and in South America, but nowhere else, is
another example.

Another very peculiar fish-type in Tanganyika is
Polypterus congicus, a member of a ‘ ganoid’ genus,
which is restricted to tropical Africa. Like Protopterus
this fish has an accessory respiratory organ, for the
organ which forms the air-bladder in many other fish
seems here to be used in respiration, perhaps because
the fish inhabits the same regions as Protopterus, i.e.
rivers whose waters become periodically foul. In very
many points of structure, however, the species of
Polypterus differ markedly from those of Protopterus.
Into these points we cannot go here, but it may be
sufficient to say that many zoologists are of opinion
that they show that terrestrial vertebrates arose from
a stock common to them and to Polypterus. Though
the lung of the dipnoi is much more like the lung of
a terrestrial vertebrate than is the air-bladder of
Protopterus, it is not believed that dipnoi are near the
line of descent of terrestrial vertebrates, largely because
of the structure of their limbs. It seems as if the limb
of Polypterus could more easily give rise to that of
a land vertebrate than that of Protopterus. We have
already considered several cases of land animals re-
acquiring modifications of structure which fit them for
life in the water, but Polypterus is interesting as sug-
gesting one way in which land animals arose from
aquatic ones. Its ancestors were certainly marine,
but were driven into fresh water by the competition
of better-organized forms which appeared in the sea.

THE ANIMALS OF LAKES AND RIVERS 203

The relatively unfavourable conditions which exist in
rivers made it necessary to develop accessory breathing-
ergans, and these apparently led the way to the acquisi-
tion of purely terrestrial characters, in the allies of the
ancestral Polypterus.

Another peculiarity of the fish fauna of Tanganyika
is the enormous number of fish belonging to the family
Cichlidae. The lake is characterized by its very large
number of fish, nearly a hundred different species
having been described. Of these, more than half
belong to the family Cichlidae, and, more remarkable
still, of the fifty-eight species of this family described
in the lake only one is known outside the lake. We have
already noted, in speaking of the Scottish lakes, that
it is not unusual to find peculiar varieties or species of
fish in lakes, but the fact that many peculiar genera
occur in Tanganyika speaks to long isolation.

The other fish of the lake are for the most part
similar to those which occur in the remaining African
lakes. For instance, there are a considerable number
of cat-fish, very usual inhabitants of equatorial lakes.
As we shall see directly the invertebrates show the same
general peculiarities as the fish—that is, we find a com-
bination of forms found in other African lakes, and of
quite peculiar forms of primitive type, the result being
to give the lake an unusually large fauna. Moore
believes that this shows that the lake had once a direct
connexion with the sea to the west, through what is
now the Congo basin. He places the rupture of this
connexion so long ago as the period called by geologists
the Jurassic, and believes that at the time of the
rupture the lake was peopled by Jurassic forms. These
in course of time evolved into the peculiar types
now found in the lake, which retain many archaic

204 THE ANIMALS OF LAKES AND RIVERS

characters. Later, the ordinary fresh-water fauna of
Africa gradually reached the lake, giving the present-
day mingling of special types with marine affinities
(halolimnic forms) and ordinary fresh-water types. This
hypothesis has not been universally accepted, and the
evidence upon which it is based depends upon detailed
anatomical points which cannot be considered here.
As the figures already given suggest, Tanganyika is
an enormous lake when compared with the lochs of
Scotland, and it is of course very inaccessible. We
cannot, therefore, hope to have the detailed information
in regard to its fauna which is available for the Scottish
lochs or for Lake Balaton. It has never been systematic-
ally sounded, so that it is not possible to distinguish
between the faunas of the different depths in detail.
The great point of interest is the occurrence of inverte-
brate animals of definitely ‘marine’ type, notably of
a fresh-water jellyfish. Invertebrates of marine type
are of course not confined to Tanganyika. We have
already mentioned their occurrence in other lakes, e.g.
in Lake Baikal, but they seem to be especially numerous
and peculiar in Tanganyika, and, as already stated,
they coexist there with ordinary fresh-water forms.
No less than fifty species of molluscs have been
described from Lake Tanganyika. Of these a con-
siderable number belong to the genera commonly
represented in masses of fresh-water elsewhere. Thus
we have species of Limnaea, Planorbis, Bithynia, Unio,
&c., the first two genera being cosmopolitan and very
old. But in addition there are a number of gastropods
of peculiar and primitive characters, with strong marine
affinities, constituting part of Moore’s ‘ halolimnic’
fauna. Some of these live at depths of 600 feet and
upwards, which is in itself an exceptional feature, for,

THE ANIMALS OF LAKES AND RIVERS 205

as we have already explained, there appears to be a
deficiency of oxygen in the depths of equatorial lakes,
and in Lake Nyasa no life has been found at depths
greater than 100 to 150 feet. The occurrence of deep-
water forms in Lake Tanganyika is perhaps associated
with the antiquity of the lake, which has given time
for the evolution of a special deep-water fauna.

The crustacea include the usual small copepods of
fresh water, and also two crabs and two prawns, all
peculiar to the lake. One of these crabs extends down-
wards from 500 to 600 feet, and is thus another example
of the deep-water fauna. Among the other interesting
forms are a peculiar polyzoon, three fresh-water sponges,
and the jellyfish (Limnocnida) already mentioned. A
closely related fresh-water jellyfish has been described
from Rhodesia, and another apparently very similar
form has been found in the tributaries of the Krishna
in India, but none of these is well known.

The protozoa of Tanganyika seem to resemble
generally those of other African lakes, and indeed those
of lakes in general.

To these accounts of characteristic lake-faunas, a
word or two may be added on the fauna of the Great
Salt Lake of Utah. Here the conditions are extra-
ordinarily unfavourable to animal life. Only one
organism can be said to have solved completely the
problem of adaptation to life in the brine. This is the
Brine-shrimp (Artemia salina), which in summer is
excessively abundant (cf. Scottish Geographical Magazine,
vol. xvii. (1901) p. 617). It is very tolerant of cold
and of variations in the salinity of the water, and the
fact that the eggs can develop parthenogenetically—that
is, without previous fertilization—enables the animals to
reproduce themselves very rapidly when the conditions

206 THE ANIMALS OF LAKES AND RIVERS

are at all favourable. During the warm season the
larvae of two flies, a Tipula and a Chironomus, also
occur in the shore waters, and a water-boatman (Corixa)
is also found in the lake. Otherwise it is singularly
barren of animal life.

Cave Faunas. As caves usually contain a consider-
able amount of water, something may be said here of
the more characteristic animals found in them. Cave
animals live in conditions which in many respects
resemble those prevailing in the ocean abysses. Thus
there is often complete darkness, and the temperature,
though it may be low, is remarkably uniform. Green
plants, also, are necessarily absent. Some of the charac-
ters of abyssal forms therefore reappear here; for
example, the eyes tend to disappear or become minute.
But as the light conditions are quite different from
those which obtain in the ocean depths, where as we
have seen (p. 172) certain rays penetrate deeply, we
find that cave animals tend to be white, instead of
showing the deep-red or black coloration found in
many animals which live in the ocean near the light
limit.

An interesting example of a cave animal is Proteus
anguinus, found in the caves of Carniola, Dalmatia,
and Carinthia. This animal is one of the permanently-
gilled amphibians already described, and reaches a
length of about a foot. The skin is without pigment,
and the eyes are below the skin. The animal remains
permanently within the water, though it rises to the
surface to breathe if the water is not quite pure.
Curiously enough, if the white body is exposed to light
pigment develops in the skin, and perfectly black
specimens can be produced by gradual exposure. In
the caves of Texas there is an allied form (T'yphlomolge

CAVE FAUNAS 207

rathbuni) of similar characters. The same type of
adaptation also reappears among the cave fish, which
have been especially studied in the Mammoth Cave of
Kentucky. One of the most interesting found there is
Amblyopsis spelaea, which is white in colour, and has
concealed eyes. To make up for the loss of the eyes
the head is furnished with a great number of tactile
papillae.

Of other cave animals the most important are crus-
tacea, especially isopods, e.g. Titanethes albus, which
has no eyes, and insects, notably beetles, e.g. Adelops
and Bathyscia, also some spiders, myriapods, and a few
molluscs, the tendency to lose eyes and the pigmenta-
tion of the skin being well marked.

REFERENCES. The first volume of the scientific results of the Bathy-
metrical Survey of the Scottish Fresh-water Lochs, conducted under the
direction of Sir John Murray and Laurence Pullar (Edinburgh, 1910),
contains several articles dealing with the fauna of lakes, as well as
detailed lists of the organisms found in the Scottish Lochs. This volume
gives also a detailed bibliography of limnological literature. The account
of the fauna of Lake Balaton given in the text is based upon the Resultate
der wissenschaftlichen Erforschung des Balatonsees, herausgegeben von
der Balatonsee-Commission der Ungarischen Geographischen Gesellschaft
(Vienna, 1906). A full account of the fauna of Lake Tanganyika is
given in The Tanganyika Problem, by J. E. Moore (London, 1903).
Other references to the same subject will be found in the bibliography
mentioned above. Many of the general works already mentioned also
discuss fresh-water faunas, and to them may be added Forel’s Handbuch
der Seenkunde (Stuttgart, 1901), and the same author’s large work on
Lake Geneva (Le Léman: Monographie limnologique, 1892-1904, Lau-
sanne), and Die Tier- und Pflanzenwelt des Siisswassers, by Dr. Otto
Zacharias and others (Leipzig, 1891). See also articles on the river
mussels, &c., of South America in Archhelenis u. Archinotis, by Hermann
von Ihering (Leipzig, 1907). Kobelt’s book (cf. p. 35) has an interest-
ing chapter on cave faunas. See also Semper’s Conditions of Existence
as they affect Animal Life (London, 1881). The blind fish of the Kentucky
cave are discussed by Putnam (Amer. Nat. 1872).

CHAPTER X
ZOOGEOGRAPHICAL REGIONS

WE have now surveyed those natural regions of the
earth described in the Introduction, and have noted
their more important occupants, and the adaptations
which these show to their particular habitats. In the
course of these descriptions it has become obvious that
what may be called the raw material upon which
adaptation has worked is not the same in all parts of
the globe. We have found that animals inhabiting
dense forest generally show certain arboreal adaptations,
such as the prehensile tail, the opposable thumb and
great toe, and so forth, found wherever the forest
occurs. But in New Guinea, for example, certain
kangaroos have taken to the trees and become arboreal ;
in Madagascar lemurs swarm in the forests; in South
America particular kinds of monkeys, the opossums,
and the sloths, as well as other animals, people the
selvas, and so on. In other words, each isolated forest,
or other area, has its own types of adapted animals.

The only possible explanation of these conditions is
the hypothesis that the specialized forms in each
separate region have been evolved from pre-existing
unspecialized forms inhabiting the region. In South
America we find fossil representatives of ground sloths.
These have long since become extinct, and the exist-
ing sloths have apparently been saved from a like
fate by the acquisition of marked arboreal characters,
which protect them from many possible enemies. In
the continent of Australia, and parts of the adjacent

ZOOGEOGRAPHICAL REGIONS 209

regions, kangaroos are abundant, and are for the most
part savana animals. But where tropical forests occur,
as in North Queensland and New Guinea, there certain
forms have taken to the arboreal life, and have thus
protected themselves from the competition of ground-
living forms no less than from the attacks of some
possible enemies.

The same statements might be made of desert
animals, of savana animals, or, generally, of any
forms inhabiting a natural region. We may say, then,
that when any stock of terrestrial animals reaches a
new region, and is there isolated from any cause, that
stock will tend to give rise to specialized groups, more
or less perfectly adapted to all the possible varieties
of habitat which the region affords. In order that
a particular stock may give rise by specialization and
differentiation to many such groups, some degree of
isolation seems to be necessary, for otherwise new
incoming stocks may colonize the vacant places before
the first stock has had time to become differentiated.
For example, in Australia almost all habitats suited
to terrestrial mammalia are occupied by marsupial
animals. Living marsupials also occur in South
America, but there they are represented by a very
limited number of types, because they have: had to
face the competition of placentals, practically absent
in Australia. Similarly, in Madagascar the lemurs are
very numerous and very diverse, being adapted for
different modes of life. Lemurs also occur in Africa
and India, but there, where they have to face the com-
petition of the more intelligent.monkeys, and the
attacks of more powerful carnivores than any which
are found in Madagascar, they are few in number,

small in size, and less diverse.
" 1404 (e)

210 ZOOGEOGRAPHICAL REGIONS

Such isolation of portions of the earth’s surface is
caused by the existence of barriers to the distribution
of terrestrial animals. Necessarily these barriers must
be supposed to have a definite date of origin, and
this origin must always be supposed to be later in :
time than that of the ancestral stock of the group
under consideration, but to precede the differentiation
of the stock into many divisions as the result of the
isolation. As an example of the origin in time of such
a barrier let us take a progressive climatic change.
The animals of temperate Europe and Asia show
a marked difference from those of tropical Asia and
of tropical Africa. To some extent this difference is
doubtless due to the existence of transverse mountain-
chains, such as the Himalayas and the Andes, but it
is also largely due to the existence of a great band of
deserts and wastes across Asia and Northern Africa, for
both the mountains and the deserts prevent intermixing.
Now there is much evidence to show that there has
been a progressive desiccation in this region within
geologically recent times. One effect has been to
favour the development of the many steppe and desert
animals of the region, but another has certainly been
to separate an originally homogeneous fauna into three
parts, and by isolating much of Africa on the one hand
and the peninsula of India with Further India on the
other, to favour evolution in these two areas, now pro-
tected from the incursion of northern forms, save on
a very limited scale. We have seen, for example, how
scantily represented in temperate Europe and Asia
are the abundant antelopes of Africa, and similarly
how the numerous deer of the north are shut out from
Africa by the, to them, impassable desert.

To the zoogeographer, however, the greatest barrier

ZOOGEOGRAPHICAL REGIONS 211

is undoubtedly the sea, for a broad channel offers an
insuperable barrier to the passage of most land animals,
save those small enough, or resistant enough, to be
carried passively by winds and waves or an animal
host. Here, obviously, we approach difficult questions.
Geological research, especially in recent years, has
familiarized geographers with the idea that great
changes in the distribution of land and water have
taken place on the surface throughout geological time.
Now to consider this distribution in its relation to
animal life and evolution from the Cambrian period
onwards would obviously be a Herculean task, though
it is one which has been attempted, e.g. by Arldt.
Without such a consideration, however, we cannot
hope to explain in its entirety the distribution of
animals at the present time. But without attempting
to do this, it is of interest to note the main features
of the distribution of the higher forms of life in the
existing continental masses. As we saw in the Intro-
duction, such a division of the globe into zoogeo-
graphical regions is primarily of interest to the zoologist
rather than to the geographer. This is especially true
when the subject is treated in detail, as in most of the
works on the subject. At the same time for historical
reasons, if for no other, the subject is too important
to be entirely omitted, and in this concluding chapter we
shall consider briefly the classical ‘ regions ’ into which
the globe has been divided by zoologists, on the basis
especially of the distribution of birds and mammals.

In dividing up the globe in this way the first point to
be emphasized is that, though the geographers are accus-
tomed to lay great stress upon the distinction between
the American and Eurasiatic continents, between
the Old and the New Worlds, no such distinction

02

212 ZOOGEOGRAPHICAL REGIONS

exists between the animals, at least of the northern
regions. We have already pointed out that the tundra
animals of the east and west show very marked resem-
blances to one another; that similarly there is con-
siderable resemblance as regards both the plants and
animals of the belt of coniferous forest in both hemi-
spheres, and that only as we travel southwards does
differentiation appear. There is much evidence, both
geological and zoological, to show that not long ago there
was free land communication between the Old and New
Worlds in the northern hemispheres, across what is
now the Bering Strait, as well as possibly across parts
of the North Atlantic. Further, though parts of what
are now the continents of Africa and of India were,
in early Tertiary times, probably separated by stretches
of sea from the land-masses to the north, and though
they are now functionally separated from the great
land-mass in the eastern temperate and frigid zones
by belts of deserts and by transverse mountain-chains,
yet, before the progressive formation of these belts of
desert, there was apparently a time when it was
physically possible for land animals to spread from
the far north of Europe and Asia to the far south of
Africa and India, and south-eastwards to parts of
the Malay Archipelago. With this great eastern land-
mass North America, as already suggested, was in
free communication, though it seems to have been long
separated from South America.

We thus have to consider as forming one great
Ream of the earth’s surface the whole Eurasian con-
tinent with parts of the Malay Archipelago, the con-
tinent of Africa with the isolated region of Madagascar,
and North America. In this great land-mass evolu-
tion, as we would expect from its great extent, has been

ZOOGEOGRAPHICAL REGIONS 213

rapid. Here we find the highest mammals and the
most specialized birds. Where the tropical forest is
dense a few primitive forms linger, and some also have
been saved by isolation, e.g. in Madagascar. But as
a general rule the primitive forms have been crushed
out of existence by higher forms. No marsupial (save
one opossum in North America) now lives within this
great area, and no monotreme. No primitive reptile,
like the Hatteria of New Zealand, points us back to an
earlier age. Here and here alone do we get the highest
of the primates apart from man—the anthropoid apes
and the dog-faced baboons. Here and here alone do
we get the highest and most specialized of the ungulates,
and so on.

For this great area various names have been pro-
posed. Thus it has been called Arctogaea, or the
Northern World, because of its mainly northern position,
though Africa extends far to the south.

In regard to the boundaries of this realm it may be
said that over much of the area they are formed by
the sea. In two regions, however, difficulty occurs.
One is in the south-east, where a boundary line has to
be drawn through that mass of islands which stretches
between Further India and the northern shores of
Australia. Wallace drew the line which separates his
Oriental (or Indian) region and the Australian region
between the islands of Bali and Lombok. A narrow
but deep strait separates these islands, and Wallace
believed that all the islands to the east of this strait
(Wallace’s line) possessed faunas with a distinctively
Australian facies, while those to the west had faunas of
Indian type. Recent detailed research has thrown
some doubt upon this statement. A mingling of
faunas certainly occurs in this region, and any hard

214 ZOOGEOGRAPHICAL REGIONS

and fast line must necessarily be of a more or less
artificial character. The difficulty is sometimes got over
by the erection of a transitional region. It is enough
for our purpose to know that such a transition exists.

The other area where a line is difficult to draw is
in the American continent. There are many striking
differences between the faunas of North America and
South America, but the two regions are now in free com-
munication. Necessarily, therefore, a certain amount
of mingling occurs. That this mingling is not greater
than it is is doubtless due to climatic differences, to
the occurrence of desert regions, e.g. in Mexico and the
west,andsoon. Here again the difficulty is usually got
over by the erection of a transitional region to include
a considerable part of the south-western United States.

The rest of the globe is sometimes included in one
Realm, forming the Notogaeic Realm, or Southern
World, and including on the one hand Southern and
Central America, with Mexico and the West Indian
islands, and on the other the continent of Australia
with Tasmania, New Guinea, the islands of New
Zealand, and some of the islands of the Malay Archi-
pelago, as well as the scattered groups of Polynesia.
This union of Australia and South America, despite
their considerable separation in space, seems to be
justified by the increasing proofs of the affinities of
their faunas, and the increasing probability that they
were connected, at a not very remote period, through
the Antarctic continent.

The Notogaeic realm so defined is characterized by
the presence of monotremes (in Australia) and mar-
supials ; by the presence of running birds and dipnoi
(these being characters shared with Africa), and by the
absence of the highly differentiated forms named above.

ZOOGEOGRAPHICAL REGIONS 215

Now it is obvious that these two realms are large
and unwieldy divisions. They are therefore in their
turn subdivided into regions. Taking the Arctogacic
realm first, we find that it may be subdivided into the
following regions: (1) the Hoxarcric, including the
whole of temperate Europe, Asia, and North America ;
(2) the ORIENTAL, including India south of the Hima-
layas, Further India, and parts of the Malay Archi-
pelago ; (3) the Erurorran, including Africa south of
the Atlas and Sahara desert. To these many would
add a fourth region for the island of Madagascar, whose
fauna is so strikingly different that it may well form
(4) the Matacasy region. But while the fauna of the
Oriental region shows many resemblances to that of
the Holarctic region, the differences being largely
attributable to differences of climate, Africa, in the
characters already mentioned, is strikingly dissimilar,
and approaches the South American and Australian
regions. Without pursuing this point further, we may
note that some authorities separate it on this account
from the Arctogaeic realm.

Again, the Notogaeic realm obviously falls into two
regions: (5) the western or NEOTROPICAL, and (6) the
eastern or AUSTRALIAN.

This ‘classification may be summed up in tabular
form as follows :

ARCTOGAEIC REALM.

(1) Hoxarctic REGION [sometimes divided into an
eastern (Palaearctic) and a western (Nearctic) sub-
region].

(2) ORIENTAL REGION.

(3) ETHIOPIAN REGION.

(4) MALAGASY REGION.

216 ZOOGEOGRAPHICAL REGIONS

NOTOGAEIC REALM.

(5) NEOTROPICAL REGION.
(6) AUSTRALIAN REGION.

Each of these regions may now be considered in
a little detail.

Beginning with the Hotarcric region, we find that
the western or Nearctic section has no primates, and
the eastern or Palaearctic only those species of Semnopi-
thecus and Macacusalready mentioned (pp.77-8). Inthe
Palaearctic section the sheep and goats are well repre-
sented, but to the west they are few (cf. p. 82). Both
sections have oxen. The antelopes are scantily repre-
sented in both, while deer are abundant. To the east,
but notin the west, camels and horses occur. Elephants
do not occur in either division, but both are rich in carni-
vores. Among the special forms it is noticeable that
the badger is peculiar to the Old World, while the
Nearctic section shares the skunk and raccoon with the
Neotropical region. In both divisions rodents are very
numerous, but this is a fact which has already been
sufficiently discussed. Among the insectivores the
mole (Talpa) is a form peculiar to the Old World,
where it extends into the Oriental region.

The bird faunas of the two regions differ from each
other somewhat markedly, but this is partly because
both regions receive many seasonal immigrants from
the hotter land-masses lying to the south of them.
The differences also chiefly affect genera or even species,
rather than orders or families. Two familiar families
of the Old World are, however, absent in the New.
Thus North America has no true starlings, their place
being taken by the hang-nests or Icteridae, to which
the rice-bird or bobolink (Dolichonyx) belongs. The

ZOOGEOGRAPHICAL REGIONS 217

flycatchers (Muscicapidae) of the Old World are also
absent in the New, their place being taken by the
tyrant shrikes (Tyrannidae). Reptiles are somewhat
scantily represented, but we have already touched
upon the abundance of tailed amphibians, the majority
of which are peculiar to the region. Something has
also been said of the fresh-water fish fauna, which is
remarkable for its development of salmon, carp, pike,
perch, sticklebacks, &c. There is nothing specially
worthy of note as regards the invertebrates.

The animals of the OrrENTAL region show so marked
a general resemblance to those of the Ethiopian, that
some naturalists would unite the two as a Palaeo-
tropical region. There are, however, certain well-
defined differences in addition to basal resemblances.

Beginning with the Primates, we find that the Oriental
region has two peculiar kinds of anthropoid apes, the
orang of Sumatra and Borneo, and the gibbons of the
Malay region. There are also many kinds of monkeys,
belonging to the genera Macacus and Semnopithecus.
The genus Cynopithecus includes only one species, the
so-called black ape, peculiar to the island of Celebes,
usually included in the Oriental region. We have
already mentioned the peculiar proboscis monkey
(Nasalis) of Borneo. There are four lemurs, including
two species of the very peculiar genus Tarsius. In the
continent of Africa there are eight species, and in
Madagascar about thirty-six species—a very interest-
ing contrast.

The Oriental region is rich in bats, having many
representatives both of the insect-eating and the fruit-
eating forms, with some peculiar genera.

Among the insectivores the very peculiar flying
lemurs (Galeopithecus) of the Philippine Islands are

218 ZOOGEOGRAPHICAL REGIONS

confined to the region, as are also the tree-shrews
(Tupaia).

Carnivores are numerous, and among the interesting
contrasts with the Ethiopian region is the presence of
bears, believed to have originated within the Oriental
region. The tiger, whose presence in India affords
another point of contrast with Africa, does not reach
Ceylon, and is believed to be a relatively recent immi-
grant into India. Like the Ethiopian region, the
Oriental one is rich in civets and their allies. True
cats of the family Felidae are also numerous.

Among the ungulates we note the presence of oxen,
the paucity of antelopes, the abundance of deer, the
presence of chevrotains, of many pigs, of rhinoceroses,
of a tapir, and an elephant, and the fact that wild
asses just reach the region.

Rodents, especially squirrels and mice, are abundant,
and include peculiar genera. The Edentates are repre-
sented by members of the genus Manis, the pangolins,
represented elsewhere only in the Ethiopian region.

Birds are very numerous, the region being especially
characterized by the number and beauty of its pheasants.
On the other hand, parrots are relatively few, com-
pared with their abundance in the Australian region.
The babbling thrushes (Timelidae) have their head-
quarters in the region, and are very numerous. Very
striking also are the bulbuls (Pycnodontidae), green
bubuls (Phyllornithidae), and the hill-tits (Leiotrichidae).
There are many reptiles; we have already spoken of
the flying lizard (Draco) and the crocodiles. The long-
snouted forms of the latter (Gavialis) are restricted to
the region, where there is one species in the Ganges.

A good many of the peculiarities of the ErHiop1an
region have been alluded to incidentally in the fore-

Fia. 45. Clouded Leopard of S.E. Asia. om the picture by W. Walls.)

(From a specimen in the Royal Scottish Museum.)

Fie. 46, The Okapi..

ZOOGEOGRAPHICAL REGIONS 219

going description, but some further details may be
added. The region is almost equally characterized by
the forms which are present, and by those which are
absent. Among the Primates it possesses the gorilla
and the chimpanzee, two peculiar forms confined, as
we have seen, to a relatively small part of the con-
tinent. Monkeys are represented by five peculiar
genera with many species, lemurs by two or three
genera and some eight species. Fruit-bats are fewer
than in India. Insectivores are peculiar and primitive.
Thus we have jumping-shrews, the aquatic potamogale
and the golden mole, the latter believed to be allied
to the curious tailless hedgehog or tenrec of Madagascar.
Among the carnivores an interesting feature is the
presence of two peculiar dogs, belonging to the two
genera of Lycaon and Otocyon, the latter having very
primitive teeth. We have already spoken of the
absence of bears, and it is noticeable that the weasel
alliance is scantily represented.

So much has been already said of the ungulates (p. 129
et seq.) that a few general statements may suffice.
Goats and sheep are practically absent, and with one
exception so are true pigs of the genus Sus, though wart
hogs, bush pigs, river hogs, and the hippopotamus occur.
There is no tapir and no camel, but rhinoceroses,
elephants, chevrotains, and several kinds of horses
occur, as well as the curious hyrax and peculiar forms
like the giraffe and okapi; antelopes are abundant.

In regard to the rodents a peculiar feature is the
absence of the true flying squirrels and their replace-
ment by the peculiar family of Anomaluridae (cf. p. 109).
The Edentates are represented by the curious aard-vark
(Orycteropus), as well as by pangolins.

The birds are not so abundant nor so beautiful as

220 ZOOGEOGRAPHICAL REGIONS

those inhabiting the Oriental region. The plantain-
eaters (Musophagidae) and the colies (Coliidae) are
among the peculiar families. Though the beautiful
sun-birds (Nectarinidae) are represented in the Oriental
and Australian regions, they are especially charac-
teristic of Africa, where they take the place of the
humming-birds of South America, which they somewhat
resemble in beauty of plumage and in habits. Parrots
are not very numerous. The true ostrich, though not
confined to the region, is very characteristic. Reptiles
are numerous, and among the fish we have to note the
presence of the dipnoan genus Protopterus, and the
curious ‘ganoids’ known as Polypterus and Cala-
moicthys. Some other peculiarities of the fresh-water
fauna have been already alluded to.

The Maagasy region or sub-region (p. 152) shows
some very striking differences from the mainland,
these differences testifying to its long isolation. Thus
there are no anthropoid apes nor monkeys, but some
thirty-six species of lemurs occur, all belonging to
peculiar genera. The insectivores are somewhat
numerous, for there are over twenty species. The
relative abundance of these and of lemurs, both help-
less forms, must be associated with the paucity and
small size of the carnivores. Of the insectivores all
save one species of musk-shrew (Crocidura) belong to
peculiar genera, and most belong to a peculiar family,
that of the tenrecs or Centetidae, which is entirely con-
fined to the island, and shows many primitive charac-
ters. The musk-shrews are widely distributed in the
Old World, and the genus is one of the three which
Madagascar shares with the adjacent mainland. The
other two are hippopotamus (with one extinct species)
and Potamochoerus, the river hog (with one species)

ZOOGEOGRAPHICAL REGIONS 221

Rodents are represented by five peculiar genera belong-
ing to the mouse section ; carnivores by some seven
members of the civet family, of which the largest is
the cat-like animal called Cryptoprocta: All the genera
of carnivora are peculiar to the island. The absence of
the true cats of the genus Felis, elsewhere so widely
distributed, is a remarkable feature.

The above include all the non-flying mammals of
Madagascar, and it will be noted that they form
a striking contrast with those of Africa. The charac-
teristic African birds, e.g. the plantain-eaters and
colies, are similarly absent, while the reptiles are very
remarkable in that some show affinity with those of-
South America rather than with those of Africa. The
chameleons present some interesting features. These
lizards occur in Africa and in India, but are more
abundant in the island of Madagascar than anywhere
else, thus showing a curious parallelism with the lemurs.

Turning now to the Notogaeic Realm we may begin
with the NEOTROPICAL REGION, which is both highly
peculiar and very rich in species. Here are no anthro-
poid apes, no dog-faced (Catarrhine) monkeys, and no
lemurs, but on the other hand a peculiar family of
broad-nosed (Platyrrhine) monkeys, and another of
small, also broad-nosed, furry forms called marmosets,
both families differing in a number of respects from all
the Old World forms. The bats are very peculiar. Here
only (with a slight extension into North America) do
we find the vampire bats, while the fruit-bats of the
Old World are completely absent, as well as another
family called the horse-shoe bats (Rhinolophidae).
Insectivores are practically absent, though in the West
Indian Islands there occur two shrews of the genus
Solenodon, believed to be related to the Centetidae of

222 ZOOGEOGRAPHICAL REGIONS

Madagascar. Of the carnivores it is only necessary to
say that bears are represented by only one species,
while raccoons are plentiful, and like the skunks are
common to this region and the Nearctic.

Very remarkable conditions are exemplified by the
ungulates. We find peculiar deer, peculiar pigs (pec-
caries), tapirs (shared with the Malayan region), llamas
(allies of the Old World camels), but no antelopes,
sheep, goats, oxen, nor horses. There are many peculiar
fossil forms, but these are beyond our range.

We have already spoken of the number and peculiar
nature of the rodents, and nothing need here be added
to what has been said on pp. 56 and 87. The peculiar
nature of the Edentates has also been pointed out, and
the presence of many kinds of opossums. In South
America there occur also other marsupials, known as
selvas (Coenolestes), which are of great interest because
they are believed to belong to the herbivorous or
diprotodont section, that of which the kangaroo is the
most familiar representative. The discovery of these
little animals in South America is of great theoretical
importance, because it was formerly believed that
the diprotodont marsupials had originated within the
Australian area, and had never occurred outside it.
Their existence, therefore, is another link in the increas-
ing chain of evidence which points to a former con-
nexion between South America and Australia.

The birds of South America are almost as peculiar
as the mammals, though, as was to be expected, many
of the characteristic forms extend also into North
America. Among the important families are the
humming-birds, the macaws (Conurinae), the toucans,
the jacamars, the motmots, the chatterers, the tanagers,
the tinamus (cf. p. 140), the curassows (Cracidae), &c.

ZOOGEOGRAPHICAL REGIONS 223

The so-called American ostrich (genus Rhea), is charac-
teristic, and is represented by several species. Among
the negative characters we may note the absence of
crows and ravens, which do not extend south of Guate-
mala. As in Australia there are many parrots. Among
the reptiles interesting forms are the rattlesnakes, the
boas and anacondas, and the lizards of the family
Iguanidae, while there is a large number of toads and
frogs, especially tree-frogs (Hylidae). Among the fish
the dipnoan called Lepidosiren is important.

Finally we must consider the markedly peculiar
AUSTRALIAN region. Here the higher or placental
mammals are absent, with the exception of bats, a few
rodents of the mouse section, the dingo or native dog
of Australia, and the pig of New Guinea, which is
probably an introduced animal. As already explained,
the region further includes the only living monotremes,
Ornithorhynchus being found in Australia and Tas-
mania, Echidna in Australia, Tasmania, and New
Guinea, and Proechidna in New Guinea only.

The marsupials are extraordinarily numerous and
very diverse, and a peculiar feature is that the more
specialized forms, e.g. the certainly modern tree-
kangaroos, occur towards the north, in Queensland and
New Guinea, &c., whereas the more primitive forms,
e.g. the Tasmanian wolf and Tasmanian devil, occur
to the south, in the island of Tasmania. This is another
link in the chain of evidence which suggests that
Australia received its original marsupials from the
south, through a connexion with South America,
rather than from the north, through the islands of
the Austro-Malayan region. We need not consider
further the marsupials of the region beyond noticing
that there are forms adapted to almost every kind of

224 ZOOGEOGRAPHICAL REGIONS

habitat, and showing curious adaptive resemblances
to the placentals. An interesting form, discovered not
many years ago, is the marsupial mole (Notoryctes),
with a curiously close resemblance to the typical mole
of the Old World.

The birds of the region are interesting and peculiar.
There are no true finches, no woodpeckers, no pheasants,
and no vultures. New Guinea is especially remarkable
for its beautiful birds of paradise, represented on the
mainland chiefly by the bower-birds. Mound-turkeys
and lyre-birds occur both on the mainland and in the
islands, and the parrots are exceedingly numerous and
very characteristic, while the related cockatoos are
almost peculiar. New Zealand, with many peculiar
birds, has some remarkable parrots, notably Nestor,
which has become carnivorous, preying upon sheep,
and the nocturnal owl parrot (Stringops) with a degene-
rate keel on its sternum, and but little power of flight.
Pigeons are also very abundant, and include the most
brightly coloured members of the order. The honey-
eaters (Meliphagidae) are peculiar to the region.

Very: striking also are the Running birds. New
Zealand has the curious little Apteryx, as well as the
extinct moa (Dinornis), Australia the emu and casso-
wary.

The reptilian fauna is less peculiar, but New Zealand
has a very primitive lizard (Hatteria, p. 228) and no
snakes. Queensland and New Guinea share with America
an abundance of tree-frogs of the family Hylidae. The
type genus Hyla occurs all over Australia, and the
presence of many members of the family in the Austra-
lian continent is the more remarkable in view of its
complete absence from Africa and India (cf. p. 118).

The fish fauna of the streams and lakes is somewhat

(‘unasn py ysoog whoy ay, ur uaunoods D wWoL7) *IAAOg OT} YIM ‘elpeaysny JO plig Jomog og, “LP ‘OL

Fia. 48. Lyre Birds with nest and eggs.
(From a@ specimen in the Royal Scottish Museum.)

ZOOGEOGRAPHICAL REGIONS 225

poor, but the presence of the very primitive dipnoan
fish Ceratodus is important. This fish, now found only
in the rivers of Queensland, lived in the far-off Devonian
period in the seas of the northern hemisphere, and is
one of the curious relics of the Australian continent.
Another is a mussel called Trigonia, now found only
off Australian coasts, but once abundant in the Jurassic
and Cretaceous seas of Europe. Still another curious
relic is the mountain shrimp (Anaspides), a primitive
type of crustacean found only in the mountain region
of Tasmania, and apparently closely related to forms
found fossil in Carboniferous and Permian beds in
Europe and North America. These old-fashioned forms
are perhaps in some respects even more striking than
the marsupials in suggesting the long isolation of
Australia, and the extraordinary differences between
its fauna and that of the rest of the world.

REFERENCES, The following are the more important books on the
subject of this chapter: Wallace, Geographical Distribution of Animals
(London, 1876); Sclater, The Geography of Mammals (London, 1899) ;
Lydecker, A Geographical History of Mammals (Cambridge, 1896);
Beddard, A Tezxt-book of Zoogeography (Cambridge, 1895); Trouessart,
La Géographie Zoologique (Paris, 1890); Heilprin, The Geographical
and Geological Distribution of Animals (London, 1887); Arldt, Die
Entwicklung der Kontinente und threr Lebewelt (Leipzig, 1907). See also
Geoffrey Smith, A Naturalist in Tasmania (Oxford, 1909), for a dis-
cussion of the origin of the Australian fauna. A very elaborate series of
plates, with a considerable amount of text and a full bibliography, will
be found in the Aflas of Zoogeography, by Bartholomew, Clarke &
Grimshaw (Edinburgh, 1911).

1404 P

APPENDIX

OUTLINE CLASSIFICATION OF ANIMALS

RzEADERS who are not well acquainted with modern zoology
may find the following outline classification useful for refer-
ence. Fuller details will be found in any of the zoological
text-books, e.g. Thomson’s Outlines of Zoology (Edinburgh),
or the more systematic Natural Histories, e.g. The Royal
Natural History (London).

Of the terms used in modern classification, the word phylum
is important. A phylum includes a group of animals
whose members are characterized by a common ground-plan
of structure, and are believed to have been descended from
a common ancestor. Thus we distinguish the members of the
phylum of vertebrates from all the varied phyla of inverte-
brates by the fact that all have at some period of life a dorsal
supporting rod called the notochord, possess a tubular nervous
system, and have their respiratory organs, whether lungs or
gills, originating from the anterior end of the food-canal.
This phylum is divided into a number of smaller units or
classes as follows:

PHYLUM VERTEBRATA.

Cuass Mammatia, including warm-blooded hairy animals
whose young are nourished by milk after birth, with three
sub-classes :

Sub-class J, Hutheria or Placentalia, the highest and most
intelligent mammals, whose young are nourished before birth
by an organ called the allantoic placenta, and are therefore
born with the form of the adult; including nine orders.

1. Primates, including monkeys, apes and man, as well as

APPENDIX 227

the more primitive lemurs, forms with usually opposable
thumb and great toe, mostly arboreal in habitat.

2. Chiroptera, or bats, insect-eating or fruit-eating forms
with the power of flight, the fore-limbs being greatly modified.

3. Insectivores, generalized and usually small animals, with
affinities both with primates and with marsupials, usually
feeding on insects, and having cusped teeth. Examples: mole,
shrew, hedgehog.

4. Carnivora, or flesh-eaters, a large order of often powerful
animals, with strong teeth, adapted either for life on land
or in water, having four to five toes. Examples: cat, dog,
bear, &c.

5. Ungulata, herbivorous animals, often of large size, with
a reduced number of toes, and teeth adapted for a vegetarian
diet. In one sub-order (Perissodactyla) the third or middle
toe tends to predominate, and may be the only one present,
e.g. horse ; in the other sub-order (Artiodactyla) the third and
fourth toes are equally developed, e.g. sheep and cow. The
elephant and the rock-conies (Hyrax) of Syria show certain
primitive characters, absent in other ungulates.

6. Rodentia, the gnawing animals, usually small, and having
characteristic chisel-edged incisor teeth, by means of which
they obtain their food. Rabbits, rats and mice are examples.

7. Cetacea, or whales and dolphins, mammals very perfectly
adapted for life in water.

8. Strenia, or sea-cows, an old-fashioned group, including
only two living forms, also adapted for life in the water.

9. Hdentata, an old-fashioned heterogeneous order, including
sloths, ant-eaters, armadillos, &c., all with relatively simple
brains.

Sub-class II, Metatheria or Marsupialia, including mammals
whose young are born imperfectly developed, and are carried
by the mother after birth in a pouch, or marsupium; comprising
two orders:

1. Polyprotodontia, carnivorous forms, with numerous in-
cisor teeth, including the opossums and the Tasmanian wolf.

P2

228 OUTLINE CLASSIFICATION OF ANIMALS

2. Diprotodontia, herbivorous forms, with only two incisor
teeth in the lower jaw, including kangaroos, the wombat, &c.

Sub-class III, Prototheria, egg-laying primitive mammals,
with one living order, Monotremata, which contains only three
living kinds of mammals, the duck-mole (Ornithorhynchus),
and the two spiny ant-eaters (Echidna and Proechidna).

Cuass Aves or Birps, including warm-blooded forms
clothed with feathers. The living forms may be divided into
two divisions :

1. Carinatae, or flying birds, with many orders, having a keel
on the sternum, and usually the power of flight.

2. Ratitae, or running birds, long-legged forms, with no keel
on the sternum and rudimentary fore-limbs, including only
a few living forms, e.g. the ostrich, cassowary, emu, &c.

Cuass Repriia on Reprizes, cold-blooded scaly animals
with five living orders, and many extinct ones :

1. Crocodilia or crocodiles and alligators.

2. Chelonia or turtles and tortoises.

3. Ophidia or snakes.

4. Lacertilia or lizards.

5. Rhynchocephalia, including only the New Zealand lizard
called Hatteria, with many primitive characters.

Ciass AMPHIBIA OR AMPHIBIANS, smooth-skinned forms with
gills in the early stages, and lungs in the later, including forms
like frogs and toads, without tails in adult life, and those like
newts and salamanders, which keep the tail throughout life.

Cuass Pisces or FisuEs, permanently aquatic forms with
gills and scales. The classification is a matter of difficulty,
owing to the vast number of fossil forms. It is sufficient to
note here that most living fishes are Teleosteans, forms with
a completely bony skeleton and a swim-bladder used as hydro-
static organ. Sharks and skates, &c., are Elasmobranchs, and
have a cartilaginous skeleton, the living forms being the
specialized remnants of a large and primitive group. As
Ganoids we may group a few living forms, very different from
each other, but sharing with the Elasmobranchs certain primi-

APPENDIX 229

tive features, though they have a swim-bladder like the
Teleosts. Such are sturgeons, bony pike, the polypterus of the
Nile, and so forth. Finally, the Dipnoi or double-breathing
fish include only three living forms (Ceratodus, Protopterus,
and Lepidosiren), interesting because they possess both lungs
and gills.

The phylum vertebrata includes in addition some other
small groups, such as the CycLosromEs or round-mouths (hag
and lamprey), the TunicaTEs or Sea-squirts, degenerate when
adult, but almost tadpole-like when young, and so forth.

Though we commonly use the term invertebrate as a con-
venient contrast with vertebrate, there are many phyla of
invertebrates not nearly related to each other. The highest
is the PHYLUM MOLLUSCA, including forms breathing by
gills, usually furnished with a limy shell, having a loose fold
of skin called the mantle, and a muscular protrusion, on which,
e.g., the snail creeps, called the foot. There are three important
classes :

1. The CepHatoropa or cuttles, marine forms in which the
shell is usually lost, active and carnivorous in habit.

2. LaMELLIBRANCHIATA or bivalves, forms like oyster and
mussel, with a double shell, usually sedentary in habit.

3. GasTEROPODA or univalves, including snails, slugs, peri-
winkle, &c., in which the shell is present or absent, and when
present has but one valve.

The very large PHYLUM ARTHROPODA includes animals
in which some at least of the segments of the body are fur-
nished with jointed appendages. There are three main
classes :

1. The Crustacna, mostly aquatic forms, with a hard coat,
two pairs of antennae, and gills, including large forms like crab
and lobster, and the small copepods or water-fleas, é&c.

2. AnTennatTa TracHEATA, including insects, millipedes,
and centipedes, and the curious worm-like Peripatus, all
breathing by air-tubes, and having one pair of feelers or
antennae. Most live on land.

230 OUTLINE CLASSIFICATION OF ANIMALS

3. ARACHNOIDEA, including spiders, scorpions, the king-
crab, &c., as well as many parasites. They have no antennae,
and the breathing organs are very diverse.

The PHYLUM ECHINODERMA includes a great number
of marine animals, of radiate form, often with limy skeletons.
The living classes comprise the sea-urchins (Echinoids), star-
fish (Asteroids), brittle-stars (Ophiuroids), sea-cucumbers
(Holothuroids), and sea-lilies (Crinoids).

The Phylum of ANNELIDS includes all the different kinds
of segmented worms, including both the .sea-worms and the
earthworms. Among worm-like types we have also many
parasitic forms, of minor importance so far as distribution is
concerned, and such free-living forms as Rotifers.

The PHYLUM COELENTERA includes a great number of
aquatic, usually marine, organisms, often of great beauty and
of great diversity. Among them we have sea-firs and their
allies, and the fresh-water hydra (HyDROMEDUSAE); the sea-
anemones, corals, alcyonarians, large jellyfish and related
forms, all of somewhat more complicated structure (Scy-
pHozoA); and finally those delicate iridescent bells which float in
the open waters of the ocean, and constitute the CreNOPHORA,
because of the delicate combs by means of which they swim.
The sponges, all aquatic and mostly marine, constitute the
PHYLUM PORIFERA, and the minute PROTOZOA, of great
importance in that many parasitic forms are included, form
a large phylum of very simple animals.

Aard-wolf, Le
vark, 135, 2

158, 169, 179-81, 206.
of yaniad 197, 204-5,

one 122, ws

Acanthephyra, 178,

Accentor, 90.

Acorn shells, 160.

Acipenser,

Acrobates, 112.

Adansonia, 122.

Addax, 128,

Adelops, 207.

Aelurus, 81.

Aepyceros, 129.

As ll a tht

5,

135-46, 145-52, 153, 182,

200-24.
Africa, Central, een
Eastern, 128 129, 1

North, 88, 128, 129, 133, 142,

210,
South, 55, 128, 129, 130.
South: western, 98, 99, 102,
106, 1) ue 119, 128,
tropical.
Aiean laces 188.
Agriophyllum, 68, 69.
Agulhas, 170.
Agutis, 110, 115.
Alactaga, 65.
Alashan, 69.
Alaska, 22,
Alca, 163.
Alces, 40.
Alcyonarians, 175.
Alders, 19.
‘Alepocephalus, 179.
Algae, 158, 159, 163, 165, 168,
* 169, 172, 173.
Alpine regions, 6
Alps, 49, 72, 74, 87, 3, 89, 90.
‘Amazon, R., 110, 192.
Amblyopsis, 207.
Amblyrhynchus ere
Amblystoma, 194, 19
Aen 21, 40, 62, 8
107, 110, 116, in
13, 18, 153, 165, 201, 224.
Central; 101; 106, 143, 147,
14; see North America,
South America.
American continent; 211,

14.
Amoebae, 188.

INDEX

Amphibians, 90, 149, 151,
166, 184, 194, 195, 206, 217.

Amphipods, ’30.

Amu Daria, 58, 59.

Anabas, 119,

‘Anaconda, 117, 223.

Anaspides, 22.

Andes te, 9, 80, 85, 87, 88,

126, 2:
Angola,
Angola, ti 68, 175, 181.
Anodon, 191.
Anomaluridae, 109, 219,
Antarctic area, 15, 21, 32,

163.
continent, 214,
Ant-eaters, 18, 94, 111.
banded, 139,
hairy, 111.
scaly, 111.
two-toed, 111.
Antechinomys, 139.
Antelope, 11, 32, 40, 60, 75,
80, 83, 84, 106, 133, 126, 1907,
128, 131, ‘132, 186, 139, 210;

African sable, 107.

harnessed, 107.

pala, 129.

saiga, 59,

water buck, 129, 130.
See apes, 98, 213,

17.

Antilocapra, 61.
Ants, 119, 142,
ee,

pteryx,
Aphrodite, 168.
Apes, 102.

black, 2a.
Aquila, 8
Arabia, a, 118, 128, 139.
Aral, Sea of,
‘Arctic America, 16,23, 25,27.

hare, 18, 20, 22, 88.
regions, 22, 31, 32, 161.
sea, 12.

waters, 33.

wolf, 21, 22.

Aretictis, 106.

Arctogaea, 213, 215.

Arctogaeic realm, 215.

Arctomys, 62, 86.

Argali, 82,

Argentine, 9, 55, 64, 113, 123,
132, 134, 135,

Argyropelecus, 178,

Armadillos, 13, 134, 135.

Artemia, 182, 205.

Artemisia, 69,

Arvicola, 87.

Asia, 11, 12, 21, 22, 32, 38, 42,

, 47, 50, 51, 54, 56, 58, 61,

62, 82, 101, 107, 119, 129, 140,
210, 15.

Central, 52, 54, 82, 83, 85,
89, 98.

Northern, 46.
South-eastern, 77, 98, 99,
106-8, 115, 117.

steppes of, BL 52, 133, 136.

tropical, 210.
Baio, itt

Ass, 61.
wild, Bee 133, 218.

Astac s, 200.
‘Atlantic, 163, 178.
‘Atlas Mts., 10, 80, 82, 126,
182, 215.
Ateles, 100.
Auk, 24, *
reat, 163.
iittle, 163.
Australia, 34, 55, 56, 102,
103, 105, 114, "1b, oe 138-
162, 191, 208, 209,
913, 214, 215, 220-5,
Eastern, 112,
North, 112.
South, 142,
West, 112.
Australian region, 135,138,

218,
Austro-Malayan region,

Axolotls, 195.
Azores, 154,

Babbling thrushes, 218.
Babirusa, 107.
Baboons, 98, 100, 124.
African, 124,
Arabian, 78.
dog-faced, 213.
Bactrian camel, 60.
Badger, 44, ae, 216,
Baffin Land, 1

Bahamas, 173,

Baikal, Lake, ae 196, 204.
seal, 182, 192, 1
Balaena, 33, 7.
Balaenoptera, 33.
Balaton, Lake, 199, 204.
Bali wen of, 213.
Baobab, 1

232,

Barnacles, 160.
Ha thy pelagic animals,174,
1

Bathyscia, 207.
ats, 5

102.

fruit, 102, 103, 217, 219.
horse-shoe, 21.
insect-eating, 217.
vampire, 103, 221.
Bay of Biscay, 165.
Bear, 44, 80, 106, 126, 222.
black, 45, 80.
brown, 45, 80.
cat, 108,
Pi eal a 45, het

alayan, 106.
. parti-coloured, 81.
sloth, 106.
spectacled, 80.
Beaver, 42, 43, 191, 193.
Bees,
Beetles, 150, 207.
water, 184,
Belgium, 59, 151.
Benguela current, 170.
Bering Strait, 16, 82, 212.
Bermudas, 154,
Bees 13, "6, 193, 216, 218,

19, 222,
of paradise, 113, 224.
rice, 16.

Bird cherry, 38.
Bison, 61.

Bithynia, 204.
Bivalves, 167,
Black Sea, 159, 180.
Blessbok, 130.

moss, 19.

Bolivia, 85, 134, 135.
Bonasa, 47.

Bonito, 177.

Bontebok, 130.

Bony pike, 185.

Borneo, 98, 104, 153, 217.
Bos, 61, 81, 127.

Bower- ‘birds, 224,
Bradypus, 110.
Brahmaputra, R., 192.
Brazil, 9, 11, 110, 134.
Brine- Lpeintos 182, 183, 205.
oe Isles, 69, 151, 152,

154
Brittle. star, 168, 184.
Bubalis, 1
Buffalo, 127.
short-horned, 127.
Bufo, 90.
Bulbuls, 218,
Bull-frog, 49.
Burma, 104. 117,
Upper, 109,
Buru, island of, 107.
Bustards, 68, 69, 140.
Butterflies, 28, 99, 119.

INDEX

Caccabis, 89,

Cactus, 123, 146, 148, 179.

Cactus wren, 141.

Calamoicthys, 220.

California, 13, 141,

Camel, 60, , 68, 85, 132,

Campylorhynchus, 141.

Canachites, 47.

Canada, 38, 40, 42, 147.

Canadian lakes, 195.

Canis, 23, 45, 67, 80, 126,

Cape Colony, 125, 133.

Cape of Good eo ie, 170,

Capercaillie, 47,

Capivara, 134.

Capra, 75, 83.

Capreolus, 41.

Capuchin pronkeys, 100.

Cariacus, 40, 1

Cariaima, 143.

Carinthia, 206.

Carniola, 206.

Carnivores, 38, 44, 67, 76, 79,
81, 124, 162, 193, 209, 216,
218, 219, 220, 921, 222,

Carp, 217.

Caspian Sea, 58, 182, 183,

Cassowary, 140, 224.
Castor, 42.

Cat- sa 199, 200, 203,
core , 67, 79, 105, 124,

Cattle, 81.

Caucasus, 63, 72, 83, 87.
Cave faunas, 206.
Cavy, 88, 134.

Cele 7e8, island of, 107,
Cemas, 83.

are pine, 38, 47, 49,

Gantotes: 152.

Centetidae, 220, 221.

Central Europe, 31.

Centrocercus, 69.

Cephalophus, 107.

Ceratodus, ue 201, 225.

Cercoleptes, 1

Cercopithecus, 99.

Certhidea, 147.

Cervulus, 107.

Cervus, 40; see Deer.

Cetacea, 162, 163, 192, 193.

Ceylon, 115, 116, ‘O18.

Chameleon, ot, 113, 116,
153, 2:

er ucie 84,

Chatterers, 222.

Chelone, 165.

Chelonia, 165, 193.

Chelydra, 193.

Chenopodiaceae, 68.

Chevrotains, 107, 110, 115,
218, 219,

water, 107.

Chickaree, 42.

Chili, 9, 85.

Chimarrogale, 78.
Chimpanzee, 86, 219.
China, 58, 77, 78.
Chinchilla, 87.
Chipmunks, 42.
Chirogale, 102.
Chironomus, 206.
Chiru, 84.
Chlamydophorus, 134.
Choloepus, 110.
Chough, Alpine, 89.
Chough-thrushes, 69.
Chrysomitris, 89.
Cichlidae, 203.

mete, 44, 102, 105, 152, 218,

palm, 106.

Cladonia, 19.

Climbing perch, 119.

Clouded leopard, 105.

Coatis, 81, 1

Cockatoos, 224.

Cockles, 167.

Cod, 161, 166.

Coelentéra, 168, 169, 185.

Coelogenys, 88, 110.

Coenolestes, 222,

Colies, 220.

Coliidae, 220.

Colobus, 99,

Bolebia f $5.

Condor, 88,

Congo basin, 203.
ygmies, 96.
‘onnochoetes, 130.

Conolophus, 148, 149.

Continental Islands, 152,

Continental Slope, 157.

Conurinae, 222

Convoluta, 168.

Corals, 181.

Corixa, 206.

Corsica, 82.

Corvus, 27.

Coryphaena, 177.

copepeds | 174, 205.
oy pu,

Grabs, 120, “167, 181, 205.

hermit, 180.

Cracidae, 222.

Cranberries, 38.

Crane, 141,

Crayfish, 196, 200.

Crested cockatoos, 114,

Cricetus,

Crinoids, 181.

Crocidura, 220.

Crocodiles, 115, 116, 194,

196, 2
Crossbill, 47, 49.
Crowberries, 19, 38.
Crustacea, 158, 160, 167, 198,

205, 207.
Cryptobranchus, 91.
Cryptoprocta, 152, 221.
Ctenomys, 134.
Ctenophora, 169.
Curassows, ‘929.

Curlews, 69.

Cuttles, 159, i 176, 184.
Cyclothone, 1

Cynomys, 64.
Cynopithecus, 217.
Cystophora, 33.

Asiatic, 106.
pampas, 132.
red, 40, 42.
Virginian, 40, 42.
Delp. inapterus, 3 34
Demoiselle crane, 7
Dendrobatinae, oa
Dendrophis, 117.
Denmark, 199.
Desert of Gobi, 58, 69.
Desman, 79, 191, 196.
Diatoms, 157.
Didelphys, 113.
Dingo, an
Dinornis, 2
Dipnoi, igs ‘201, 202, 214,
220, 223, 935,

Tepeptagant marsupials,

0, li
Dog. 106, 125, 219.
TOEn hunting, 126.
imo, 23

Talia wild, 80.
Dolichonyx, 47, 216,
ge 159, 163, 177.
blind, 192.
fish-eating, 177.
fresh-water, 192, 196.
Dormouse, 37.
Draco, 93, 116, 218.
Dryophis, 117.
Duck, 69.
Duck-mole or Ornitho-

Ue ne 191, 223.

pute on ie 107, 130.
eee Gate, 58.

Eagle, Golden, 68.
hawk, 88.
Echidna, 223.
Echinoderms, 168, 175, 178,
181, 184.
Ecuador, 85, 145.
iflontates, 110, 134, 218, 219,

Kel 176, 183, 199.
mud, 195,

INDEX

Hider-duck, 24.
Eland, 128,
Maley hant, 108, 109, 216, 218,

Elk, 40, 41,
Elymus, 64,
Fa, 67, 140, 224.
Em: 94,

England. 200.

Equus, 59.

Eremias, 70.

Erethizon, 43.

Ermine, 23, 26, 46.

Eryx, 70.”

Eskimo, 28.

Epona region, 215, 217,

rihenae 117.

Euphorbia, 122, 146.

rk ta continent, 40,

Hurope, 13, 21, 50, 63, 68, 72,
4,88, 89, 116, 130, 210,

Central, 78, 194.

South, 89,194,
European Russia, 16,
Exocoetus, 177.

Falco, 27,
Felidae, 218.
Felis, 22, 44, 79.

ciel ie

snow.

Firth = ‘orth, 164, 186,

Fish, 119, 120, 152, 178, 196,
199, 223.

cave, 207.

deep-sea, 174, 178.
flying, 177.

fresh-water, 118, 151, 183,
185, 200, 201, 203, 217; 224°

Flies, 119, 208.

Flycatchers, 217.

pane lemur, 93, 104, 217.

Fox, 37, 44, 45, 80.

Arctic, 23, 26.

corsac, 67.

long-eared, 126.

Franz‘Josef Land, 16.

Friendly Islands, ‘67.

Frog, 90, 93.

fishing, 166.

flying, 93, 117.

tree, 49, 118, 224,

Fruit-pigeon, 115.

Galago, 93, 102.
Galapagos Islands, 145-54,

Galeopithecus, 104, 217.

233

Ganges, R., 189, 192, 196,

Gannet, 163.
Ganoids, 185, 192, 200, 202,

Gastropods, 167, 175, 190.
Gavialis,

Gazelle, 59, 84, 129.
Gecinus, 47.

Gecko, a 149,

Geese, 69

Gemsbok, 128, 129, 130.
Geospiza,. 147.
Gerfalcon, 27.
Germany, 63.

Gibbon, 98, 99, 217.
Girate, 122, 129, 131, 136, 139,

Glass crab, 176.
Globigerina, 181.
eons. 24, 4.

u, 130.
Goat, a 40, 75, 76, 79, 83, 84,
216.

Pecried wild, 83.
Rock: y Mts., 4,

Gobies, 160, 166, 199.
Golden mole, 219,
Ganon, 178.

Gopher, 62.

Goral, 83.

Gorilla, 98, 219.

Graculus, ’s9
Grasshopper, 71.

Great Salt Lake, 182, 183,

Greece, 71.
Greenland, 16-29, 32.
Grinnell Land, 22, 23,
Grosbeak, 47.
Grouse, black, 47.
Canadian, 47.
hazel, 47.
Pallas’s sand, 68, 69.
red, 151.
ruffed, 47.
willow, 151.
Grus, 70.
Guanaco, 85.
Gudgeon, 199.
Guillemots, 24, 163.
Guinea-fow], 115.
Gulf of Mexico, 55.
Stream, 170.
Gulo, 24, 46.
Gypaétus, 88.

Haddock, 166.
Hairmoss, 19.
Halicore, 162.
Halobates, 176.
alolimnic fauna, 204,

Hamster, 66.

Hangnests, 216.

Hare, American, 42.
Arctic, 18, 20, 22, 88; see
also Polar hare.

234

Hatteria, 213, 204,
Hedgehog, tailless,152, 219.
emitragus, 83,
ermit crabs, 166, 180.
erpetodryas, 117.
erring, 166, 182.
esperomys, 146,
eterocephalus, 134.
ill-tits, 218,
imalayas, 10, 72, 76, 77,
80-9, 210, 215.
Hippopotamus, 219, 220,
jort, 161, 171, 172, 174,
oatzin, 115.
Hog, river, 219, 220.
wart, 219.
Holarctic region, 215, 216.
Honey creepers, 147.
Honey eaters,
Horses, 60,61, 65, 68, 108, 146,
216, 219,

Hubara, 68.

Humuming- tae gh ig 222.

Hungary, 55, 58, 1

Hyaenas, 125, 126,

Hydra, 191, 196.

Hydrochoerus, 134.

Hydromedusae, 189, 190,

Hydrophis, 165.

Hyla, 224.

Hylidae 49, ae 223, 224,
Hyperodon, 17 6.

Te enon, 136,

Hyrax, 86, 219.

iapjavtariusjaul

Ibex, 83.

Iceland, 16.

Icteridae, 216,

Ictonyx, 127,

Iguana, 116, os

India, 77, 80, 00, 104-7,
109, 11 114, eat 153.
India, Further, 210, 213,

Indian Ocean, 118, 148, 155,
162, 165,

Indies, West, 116, 221.
Indo-Malayan region, 112,

Indus, R., 192.
Insectivores, G. 79, 152, 191,
193, 216, 2 aii, 2 9, 220, 221.
Insects, 119, i
leaf, 130,
Ireland, 151.
Isopods, 207.

Tynx, 47.

arma 114, 222,
Jackal, 125,
Jaguar, 105, 108.

INDEX

Japan, 155,
Java, 155,
Jays, 47, 49.
Je ryfish, 169.
fresh-water, 204, 205.
J ee 64, 65, 66, 124, 133,

136, Nf
five-toed, 65.
Jumping mouse, 62, 138.
Jumping Sbrem. 124, 219.
Jungle fowl, 115.
Juniper, 38.

ioiaher aad 123, 124,
128, 129, 131.
Kamchatka, 82,
Kangaroo, 56,
112, 128, 135-8,

222,

five-toed, 136.

giant, 136.

Tat, 136.

tree, 111, 135, 223.
Kansu, 78.

Kea, 88.

Khingan Mts., 57.
Kiang, 60, 85,
Kilimanjaro, 86.
King Oscar Land, 16, 21, 22,

Einkajou, 94, 106.

iz, 59.
Kittrwarke, 24, 164.
Kiwi, 140, 150.
<lipspringer, 84, 129, 130.
oala, 113,
<ordofan desert, 122.
Cudu, 128,

Kuku Nor, Lake, 77.

64, ag 111,
209;

Labrador, 16, 170.

Lacerta, 90.

Ladak, 79.

Laemmergeier, 88.

Lagidium, 87.

Lagomys, 66,

Lagopus, 20, 25, 151.

Lagostomus, 134,

Laminarian zone, 158.

Langur, 77, 99; see also
Semnopithecus.
Himalayan, 78.

Tibetan, 78, 80.

Larch, Siberian, 38.

Lark, ‘Siberian, 69.

steppe, 66.

Lelotaehidas, 218,

*Rretle Si 20, 22, 23, 27, 66.
Arctic, 87,

Norwegian, 22.

Lemur, 95, 101, 102, 152, 153,
217, 219, 220, 221,

flying, 93 104, 217.
mouse, 162,

ring-tailed, 102.

Leopard, 105.

Lepidosiren, 118, 201, 223,
Leptocephalus, 176.
Leptonychotes, 161,
Lepus, 22, 88.

Lerwa, 89.

Leven, L., 197.

imnaea, 204.
dimnocnida, 205.
Limpet, 160.

uinota, 20.
7}
Ld.

on, 105, 125, 129.
thocranius, 129.
Littoral animals, 161, 167,

annelids, 168,
area, 170.
birds, 163.
fauna, 196, 199.
fishes, 166, 179.

Lizard, 69, 90, 94, ale, 117,
142, 148, 153, "901; 2:

flying, d18.

marine, 149, 165.

Llamas, 132, 922,

Lobelias, 72,

Lobodon, 162.

Lobster, 167.

Locusts, 71, 123.

Loggerhead turtle, 165.

Lombok, 213.

Lophyrus, 50.

Loxia, 49.

Lumpsucker, 160, 166.

Lung-fishes, 188.

Lutraria, 167,

pyran, 126, 219.

Lynx, 3 44, 78, 80.

Lyre- binds bot 224.

Lyrurus, 47.

Macacus, 77, We 216, 217.
Macaques
Tibetan,
lacaws, fe 8 222,
lackenzie R., 21.
fackerel, 177.
acrorhinus, 162,
acroscelides, 124,
acrurus, 180,
Madagascar, 101, 102, 105,
117, 152, 153, 155, 208° 209,
212) 213, "915, ‘17, 519, 920-2
Magpie, 49.
Malagasy region, 215, 220,
Malay ‘Archipelago, 212,
214, 215.
Peninsula, 102, 104, 109,
region, 116, 158, 262, 217,

Mammoth, 22.

Mammoth Cave of Ken-
tucky, 207.

Manatee, 162, 163, 192, 196.

Manis, 111, 21 is.

Maral stag, 40.

Markhor, 83.

Merniosess i, 101, 221.

BSSSSE4

Marmot, 46, 62, 86, 87.

Alpine, 63 , 87.
Himalayan, 87.
Prairie, 64.

Rocky Aeantais 87.
Marsupials, 56, 123, 135, 138,
209, 214, 239, 293, 295,
Marsupial mole, 224.
Marten, 44, 45.
Mediterranean, 10, 159, 163,
Medusa, 178.
Medusoids, 190.
Meerkat, 125.
Melanocorypha, 69,
Meles, 4
Melichacidac, 224.
Melursus, 106,
Mephitis, 44.
Mesonekton, 177.
Mesoplankton, 177.
Mesopotamia,” 139.
Mexico, 43, 100, 110, 194, 214.
Mice, 146, 218.
‘igheael § Sars Expedition,
Microchiroptera, 103.
Micropus, 90.
Minx, 44, 46/
Moa R., 58, 195.

Moltses, Cy 181, 184, 197,

Molson, 142,"

Mongolia, 41, 66, 82.

Mongolian gazelle, 59,

Mongoose, I:

Monkey, 11, 76, “77, 78, 93, 94,
Ne -100, 102; 105, 153, 209,

broad-nosed, 100, 221.
howling, 100.
proboscis, 217.
spider, 100.
thumbless, 93, 100.
Monodon, 33.
uonetene. 191, 214.
Monticola, $0.
ontifringilla, 89.
Morar, L., 197.
Moschus, 41, 85.
Mosquitoes, 27, 28, 119.
Motmots, 222.
Mound turkeys, 224.
Mountain ash, 38.
Mountain shrimp, 225.
Mt. Kenia, 72.
Mouse, 221, 223.
Mozambique, 133.
Mudskippers, 118.
Srna, 107.
Mus, 146,
Muscicapidae, 217.
Musk deer, apf 85.

ox, 17, 20, 21, 23, 81.
shrew, 230.
Musophagidae, 220

bo

INDEX

Musquash, 4
Museel, 1 167, “ia, 196, 197,

Mustela, 23, 45.
Mya, 30, 167.
Mycetes, 100.
Myodes, 22.
Myogale, 79.
Myopotamus, 134.
Myrmecobius, 139.
Myrmecophaga, 111.
Myriapods, 207. '
Mysis, 198, 199.

Nansen, a 26.
Narwhal, 33

Nasalis, 217.

Nasua, 81, 106.

Nearctic "region, 215, 216,

Nectarinidae, 220.
Necturus, 195.

Nekton, 176.
Nemorhoedus, 83,
Neotropical region, 216,

Nereis, 168,

New Guinea, $6, 111, 113,
208, 209, 214, 223, 224°

New Siberian Islands, 16.

Newts, 195,

New arte » 102, 140, 150,
212, 214, 22:

Nilgiri Hille, 83,

Nisaétus, 88.

Noctiluca, 175.

Nordenski6ld, 17, 20, 31.

North ‘America, ‘1, 12, 18,
34, 38, 41-51, 56, 61, 64, 87,
170, 183, 1947 213, 215, ‘216,
221; 2297 995.

North ‘Atlantic, 163,176,212,

North Pacific, 162,

North Sea, 157, 161.

Northern ‘Asia, 15, 25, 27.

Northern Russia, 19.

Norway, 72.

Norwegian Sea, 161, 170.

Notogaeic realm, oid, 215,

Notoryctes, 224,

Nova Zembla, 16, 22, 27, 31.
Nucifraga, 49.

Nuterac ore Hs 49,
Nyasa, L.

Nyctea, 3

Oceanic islands, 151.
Okapi, 107, 219.
Ommatophoca, 161,
Opisthocomus, 115.
Opossum, 95, 113, 218, 222.
Orang, 98, 217.

Orca,

235

Orcyteropus, 185, 219,
Oreotragus, 84,
aa ae 215, 216,

omnithorhynchus, 191, 223,
Oryx, 12

Ostrich, er 13, 140, 220, 223.
Otaria, 161.

Otis,

Otocy on,

Otter, 1B Tah, 193, 196.
Ounce, 79.

Ovibos, aL.

Ovis, 75, 8

Owl, 27, if.

Owl’ parrot,

oni 128, nes 218.

Pacas, 88, 110.
Pate Ocean, 118, 145, 162,

Palaearctic region, 216.
Palaeotropical region, 217.
Palalo worm, 167.

114,
Parrots, 94, is 218, 220, 223,

2

Partridge, 140.

Greek, 89.

snow, 89.

wood, 115.
Patagonia, 55, 85, 134.
Peccaries, 107, 117.
Pedetes, 133.
Pedioecetes, 69.
Pelagic animals, 157,

159, 161-7, 175-80, 198.

Pelagic fish, 17

larvae, 167, 179.

region, 170,
Pelagothuria, 178.
Pelea,

Pelias, 90.
Penguins, 163.
Perch, 199, 217.
Periophthalmus, 118,
Perisoreus, 49.
Periwinkles, 167.
Persia, 58.
Persian gazelle, 59.
Petauroides, 112.
eden 112,
Petrels, 24.
Phyllirhée, 175.
Picas, 66, 67,
Pichiciago, 134,
Pigeon, 224.
Pig-footed ‘bandicoot, 138,

158,

236

Pigs, 107, 108, 146, 147, 218,
219, 222, 298.

bush, 107, 219.

wild, 7

Pike, 199, ‘O17.

Pine grosbeak, 49.

Pine-marten, 45.

Pine saw-fly, 50.

Pinicola, 49,

Pipefish, 166.

Pisidium, 198.

Phacochoerus, 107.

cc ae aa 93, 94, 112, 113,

flying, 112.
Phascolarctos, 113,
Pheasant, 68, $9, 115, 218.
Philippine Islands, 04, 217.
negritos, 96.

Phoca, 33, 192.

Pholas, 167,
Phrynocephalus, 69.
Phrynosoma, 142.
Phyllodactylus, 149.
Phyllornithidae, 218.
Phyllostomatidae, 103.
Physeter, 176.

poy enlankton, 158, 173.
Plaice, 166.

Plankton, 174, 176, 177, 180,

Planorbis, 204.
Plantain-eaters, 114, 220.
Planulae, 120.
Platanista, 192.
Plateaux, 43, 75, 78, 82.
Ee tyrthine inonkeys, 100,

PBSEORERE: 26.

Plovers, 67.

Podoces, 69.

Poecilogale, 127.

Polar bear, ‘04, 32.

hare, 42; see also Hare,
Arctic,”

Polecat, 46, 127, 184.

Polychaete worms,
189, 190.

Polynesia, 214.

Holy phenes) 185, 202, 203,

Polytrichum, 19,
Pond snails, 184.
Porcupine, 43.

Canadian, 42, 110.

tree, 43, 94, 110.
Porpoises, 159, 177.
Potamochoerus, 220.
Potamogale, 219,
Potorous, 136,
Prairie-chicken, 69.

-hen, 69.
Prawns. 178, 184, 205.
Prej evalski’s horse, 59, 60.
PERS, 76, 78, 94, 97, 124,

Sreenttaad, 223.
Prongbuck, 61.

a

INDEX

Proteles, 125.
Proteus, 206.
Protopterus, 118, 201,” 202,

Protozoa, 170, 174, 198, 205.
Ptarmigan, 19, 2, 23, 25,
Pteromys, 109.
Pteropods, 177.

Pudua, 85.

puting 7 163, 164.

Puma, 44 79, 105.
Punjab, ib, 82.
Pycnodontidae, 218.
Evemy binpepots ea ih 153,
Pyrenees, 3, 78, .
Pyrosoma, 175.
Pythons, 116.

Quagga, 182, 133.
Queensland, 11, 112, 118,
136, 201, 209, 223, 994°

Bapeoane, 81, 106, 126, 216,
rere 49, 90.

Rangifer, 21;
indeon

see also

Rat.

Retleinake 223,

Raven, 27.

Realms, zoological, 212.

Redpoll, 20.

Red Sea, 170.

Reindeer, 17, 20, 21, 26, 41.
moss, 18.

Reitbok, i

Reptiles, 90 115, 142, 147,
130, 151 » 16, 193, 196, 218,

2
Rhacophorus, id 117.
Rhea, 67, 139, 223.

Rhebok, 84, 129.
Rhinoceros, 108, 132, 218,

219.

Rhinolophidae, 221.

Rice-bird, 147, 316.

River-hog, 153.

Roach, 199, c

Rock coney, 86.

Rock lobster, 176.

Rock thrush, 90.

Rocky Mountains, 44, 79,
80, 82, 84.

Rodents, 38, 60, 64-7, 78,
86, 89, 108, 110, 112, 124,
125, 133-6, 152, 191, 193,
216, 218, 219, 221-3.

Roedeer, 41.

Rorqual, 33.

ety ers, 184, 187, 198, 200,

Running birds, 224,
Rupicapra, 84.
Russia, Southern, 55, 58,

Russian Turkestan, 48.
Ruwenzori, 72.

Sage-brush, 69.

cock, 69, 70.

Sagitta, 175.

Sena, ie 80, 106, 118, 121,
126, 131, 215.

Saiga ene. 59, 84.

Salamander, 90, 91, 194, 195.

Salmon, 183, 217.

Salpa, 175.

Sambar, 107.

sondpiper, ri

Send -rat,

-snake, 40

Sandwich Islands, 154.

Sarcorhampus, 88.

Sardinia, 82.

Sargasso Sea, 159, 173.

Saxicava, 30, 1 7.

Sayansk divide, 41.

Scandinavia, 16, 151.

Scincus, 142,

Sciuropterus, 42,

Sciurus, 42,

Scorpion, 120,

Scotland, 72.

lakes of, 197, 208, 20

Seals, 24, 30, 32, 33, 16i, 163,
182, 193,

Seal, bearded, 33.
crab- -eating, 161.
crested, 33.

earless, 31.

fur, 161, 163, 182.
Greenland, 33.
ringed, 33, 192.
Ross’s, 161.
Weddell’s, 161.
Sea-anemone, 169, 189.
-butterfly, 175, 177.
-cow, 162.
-elephant, 162.
-fir, 169, 189, 191.
-horse, 166.
-leopard, 162,
-otter, 162.
-pen, 169.
-snake, 165.
-squirt, 166.
-urchin, 31, 168, 181, 184.
Sebastes, 178.

Sechwan, 78.

Secretary bird, 140, 141,

Selvas, 222.
Semnopithecus, 77, 216,217.
Senecios, 72.
Seriemas, 141, 142.
Serow, 83, 84,
Serpentarius, 140.
Serpula, 168.
eee 39, Hf, sft we
9, 128,

Barb: ary, on
Bighorn, 82.

‘

79, 80,

Shell-fish, 166,

Shrews, 152, 221,

Alpine, 78.

fying 112,
en swimming,

jumping, 124, 219,

jap 250.

Tibetan mole, 79.

tree, 104, — ee.

water, 191, 1

ere 13, 16 4, 45, 46, 49,

sinerien fir, 38.
Silurus, 199,

Sind, 82.

Siren, 195.

Sirenia, 162, 192, 193.

Skua, 2:
Skunk, 4a, 127, 216, 222,
Sloth, 11, 13, '93, '95, 110,

ground, 208.

Snail, fresh-water, 190, 197.

Snake, 116, 148, 149, 153.
grass, 194:

sea, 165.

tree, 117.

whip, 117.

wood, Pe

Snipe, 6

sre ehiag: 19, 26.
-cock, 89.

Snowy owl, 27

Solenodon, 221.

Solitaire, 150.

Somaliland, 125, 134.

Sorex, 78.

South America, 9, 12, 34,
48, 55, 56, 80, 85, 87, 88, 93,

130-4, 139-47, 14
202, 208, 209, 214, 215, 220,
921, 2997 993.
Southern England, 59.
France, 59.
penepE ete, 15.
Spain, 73.
See eee 62.
Sphagnum, 19.
Spiders, 120, 207.
Spitsbergen, 16, 21, 22, 23,

Sponges, 169, 174, 175, 180,
fresh-water, 205.
epebiail s cuscus, 112.

ee sy 129,

eon ‘owl, 1

Panicrel, is, Gr, 42, 43, 62,
105, 1

flying, . 6, 93, 104, 108,
109, . .

pygmy flying, 112.

sugar, 112.

INDEX

Starfish, 31.
Starling, 216.
Steinbok, 129,
Stenorhynchus, 162.
Sterlet, 200.
Stickleback, 217.
Stoat, 26, 44.
Strepsiceros, 128.
Strepsiptera, 28.
Stringops, 224,
Struthio, 139.
Sturgeon, 185, 200,
Sudan, 121, 124.
Sumatra, 98, 217.
Sun-birds, 220.
Sunfish, 174.
Suricata, 125.
Sus, 39, 219.
Suslik, 62,
Sverdrup Expedition, 16,
18, 21, 26.
Switt, ‘Alpine, 90.
Switzerland, 63, 74.
Syntheres, 110.
Syr Daria, 58.
Syria, 86, 130, 139.
Syrrhaptes, 68.

Tahr, 83.
Tailless hedgehog, 152, 219.

Takin, 83.

Talpa, 216.

Tamias, 42.

Tanagers, 222.

Tanganyika, Lake, 182,
200-5.

Tapir, 85, 108, ae 220, 222.

Tardigrada, 19

Tarim basin, 38

Tarsier, 102.

Tarsipes, 112.

Tarsius, 102, 217.

Tasmania, 102, 214.

Tasmanian devil, 223.
wolf, 223.

Taxidea, 46.

Tench, 199.

Tenrec, 219, 220.

Teratoscincus, 70.

Terebella, 168.

Termites, 120, 125, 135, 142.

Terns, 24.

Tetrao, 47.

Tetraogallus, 89.

Tetrastes, 47.

Texas, 141, 296.

Thalassochelys, 165.

Thynnus, 177.

bey 57, 73, 74, 77, 80-4,

87-9.
Tichodroma, 89.
Ticks, 120, 130.
Tiger, 44, 67, 105, 218.
Timelidae, 318.
Tinamus, 140, 222.
Tipula, 206.
Titanethes, 207.

237

Toad, 90, 223.

horned, 142,
Tomopteris, 175.
Tortoise, 165.
giant, 147, 148, 150, 154.

pond, i94.

water, 184, 196.
Toucans, 114, 222.
Tragelaphus, 107, 128.
Tragulus, 107.
Tree-creeper, 37.

-frog, 37, 49, 228, 224.

-heath, 72,

-shrew, 104, 105, 218.
Trichechus, 32.
Trigonia, 235.
Trigonocephalus, 7
Trimeresurus, We
Trochospheres, 190.
Tropidonotus, 194.
Tropidurus, 149.
Trypanosomes, 131.
Tsetse flies, 119, 130, 131.
Tucotucos, 134.
Tunicates, 175.
Tupaia, 104, 218.
Turbot,

Turkestan, 70.

Turtle, 164.

‘American snapping, 193.
reen, li
awksbill, 165.

leathery, 165.

loggerhead, 165.

marine, 193, 194.
Tympanuchus, 69.
Typhlomolge, 206,
Tyrant shrikes, 217.
Tyrol, 89.

Ungulates, 35, 56, 60, 61, 65,
75, 76, 81, 84, 86, 89, 97, 106,
108, 110, 123, 127, 130, 131,
133, 136, 137, 152; 153, 213,
218, 219, 292°

Unio, 204.

United pelea 55, 113, 142,
147, 195.

Upper eent, 83.

Uropsilus, 79.

Ursus, 24, 45, 80, 106.

Victoria, 112.
Victoria Land, 187.
Vicuna, 85.
Viper, 90.

tree, 117.
Viscacha, 64, 134.
Viverridae, 105.
Voles, 87.
Vultures, 76.

Wallabies, 123, 135.
Wallace, 146, 147, 150, 151,

152, 154.
Wallace’ s line, 213,

238

Wall-creeper, 89.

ween 24, 30, 31, 32, 101,

162.
Wapiti, 39, 40, 42.
Wart-hogs, 107.
Water bears, 198.
beetles, 197.
boatmen, 197, 206.
Tn a 107,
fleas, 16 7, 198,
mites, oa 198,
shrew, 191, 168,
spider, 184.

vole, 191, 193.
Weasel, 44, 46, 127, 219.

INDEX

Welwitschia, 124
West Indies, 147, 182, 214,

221,
Whales, 31, 159, 162.
bottle-nose, 176.
Greenland, 33,
killer, 32, 34.
sperm, 176
Snolehone, 30, 162, 176,
white, 34,
Whelks, 167.
Whortleberries, 20, 38,
Wild cat, 44,
boar, 39,
dog, 44.

Willows, 19.

Wolf, 23, 37, 44, 45.
Wolverene, 46. .
Woesdpeokers) 37, 47, 109,

een, 47,
spotted, 47.

t ree-toed, 48,

Wood wasp, 50.

Wryneck, 47.

Yak, 81, 82.

Zebra, 60, 132, 133.

OXFORD: HORACE HART M.A.
PRINTER TO THE UNIVERSITY